Guidelines for Analyzing Curved and Skewed Bridges

Transcription

Guidelines for Analyzing Curved
and Skewed Bridges and
Designing Them for Construction
FINAL REPORT
August 15, 2010
By Daniel Linzell, Abner Chen, Mohammad
Sharafbayani, Junwon Seo, Deanna Nevling,
Tanit Jaissa-Ard and Omar Ashour
The Thomas D. Larson
Pennsylvania Transportation Institute
COMMONWEALTH OF PENNSYLVANIA
DEPARTMENT OF TRANSPORTATION
CONTRACT No. 510602
PROJECT No. PSU-009
Technical Report Documentation Page
1. Report No.
2. Government Accession No.
3. Recipient’s Catalog No.
FHWA-PA-2010-013-PSU 009
4. Title and Subtitle
5. Report Date
Guidelines for Analyzing Curved and Skewed Bridges and Designing Them
for Construction
August 15, 2010
6. Performing Organization Code
7. Author(s)
8. Performing Organization Report No.
Daniel Linzell, PhD, PE, Abner Chen, PhD, Mohammad Sharafbayani,
Junwon Seo, PhD, Deanna Nevling, PhD, Tanit Jaissa-Ard, and Omar Ashour
LTI 2010-18
9. Performing Organization Name and Address
10. Work Unit No. (TRAIS)
The Thomas D. Larson Pennsylvania Transportation Institute
The Pennsylvania State University
201 Transportation Research Building
University Park, PA 16802-4710
11. Contract or Grant No.
510602, PSU 009
12. Sponsoring Agency Name and Address
13. Type of Report and Period Covered
The Pennsylvania Department of Transportation
Bureau of Planning and Research
Commonwealth Keystone Building
th
400 North Street, 6 Floor
Harrisburg, PA 17120-0064
Final Report
8/16/2007 – 8/15/2010
14. Sponsoring Agency Code
15. Supplementary Notes
COTR: Tom Macioce, 717-787-2881, tmacioce@state.pa.us
16. Abstract
Although the use of curved and skewed bridges continues to increase steadily throughout the United States, certain aspects of
their behavior during construction and while in service still are not well understood. The effects of design, fabrication, and
construction on the geometry and load distribution in a curved or skewed bridge system are areas in which further study and
understanding are required. This project utilized remote acquisition capabilities for instruments on two structures in the Interstate
99 corridor: a horizontally curved, steel, I-girder bridge, and a skewed, pre-stressed, concrete bridge. Data obtained from these
structures were examined and the numerical model accuracy for curved and skewed, steel, I-girder bridges and select appropriate
model types and software was investigated. Parametric studies were undertaken on a group of representative curved and skewed
steel bridge structures to numerically examine the influence of specific variables on behavior during construction. Results enabled
the identification of preferred erection sequencing approaches. Among other results, girder vertical deflections were decreased
when paired-girder erection methods were used and paired inner erection was preferred for structures with severe curvature.
Erection methods examined herein did not show appreciable influence on skewed bridge behavior. Drop-in erection would be an
acceptable approach for either curved or skewed bridges. The findings and the numerical modeling from the parametric studies
formed the basis for suggesting possible modifications to relevant PennDOT publications. Web out-of-plumbness did not cause
appreciable bridge deflection and stress increases when the out-of-plumbness was within the limit (1%) specified in the Structural
Welding Code. Exceeding the 1% limit of the web out-of-plumbness can result in slightly higher deformations and stresses. The
use of temporary construction shoring can significantly reduce girder deflections, leading to a more constructible condition.
Inconsistent cross-frame detailing increased vertical and radial deflections in curved bridges and lateral deflections in skewed
bridges. Replacing solid plate diaphragms in skewed bridges slightly increased deformations but did not severely affect cross frame stresses. The applied temperature change did not have an appreciable impact on overall bridge deflections and stresses for
all of the radii, skew angles and cross-frame spacings studied.
17. Key Words
Bridge design, curved and skewed bridges, behavior, construction, numerical
modeling, guidelines
18. Distribution Statement
No restrictions. This document is available
from the National Technical Information Service,
Springfield, VA 22161
19. Security Classif. (of this report)
20. Security Classif. (of this page)
21. No. of Pages
Unclassified
Unclassified
421
Form DOT F 1700.7
(8-72)
22. Price
Reproduction of completed page authorized
This work was sponsored by the Pennsylvania Department of Transportation, the Mid-Atlantic
Universities Transportation Center, and the U.S. Department of Transportation, Federal Highway
Administration. The contents of this report reflect the views of the authors, who are responsible
for the facts and the accuracy of the data presented herein. The contents do not necessarily reflect
the official views or policies of the Federal Highway Administration, U.S. Department of
Transportation, the Mid-Atlantic Universities Transportation Center, or the Commonwealth of
Pennsylvania at the time of publication. This report does not constitute a standard, specification,
or regulation.
Table of Contents
1
INTRODUCTION .............................................................................................................................................................. 1
2
UPDATED LITERATURE SEARCH .............................................................................................................................. 1
3
DATA ACQUISITION SYSTEM MAINTENANCE ...................................................................................................... 1
3.1 Instrumentation Summary ......................................................................................................................................... 2
3.1.1
Structure #207 ................................................................................................................................................. 2
3.1.2
Structure #314 ................................................................................................................................................. 4
3.2 Data Collection ............................................................................................................................................................ 4
3.3 Data Reduction and Results ....................................................................................................................................... 4
3.3.1
Structure #207 ................................................................................................................................................. 4
3.3.2
Structure #314 ................................................................................................................................................. 6
4
NUMERICAL MODELING ............................................................................................................................................. 9
4.1 Parametric Study Modeling Procedure ................................................................................................................... 10
5
PARAMETRIC STUDIES .............................................................................................................................................. 10
5.1 Parametric Structure Selection and Design ............................................................................................................ 10
5.1.1
Initial Bridge Parameter Identification and Final Structure Down Selection ................................................ 10
5.1.2
Structure Design and Final Proportions ........................................................................................................ 18
5.2 Girder and Cross-Frame Erection Sequencing ...................................................................................................... 21
5.2.1
Parametric Studies - Curved.......................................................................................................................... 22
5.2.2
Parametric Studies - Skewed ......................................................................................................................... 81
5.2.3
Final Results and Discussion ........................................................................................................................ 88
5.3 Web-Plumbness ......................................................................................................................................................... 88
5.3.1
Parametric Studies......................................................................................................................................... 89
5.3.2
Results and Discussion.................................................................................................................................. 90
5.4 Temporary Shoring Placement and Settlement Effects ....................................................................................... 106
5.4.1
Parametric Studies....................................................................................................................................... 106
5.4.2
Results and Discussion................................................................................................................................ 114
5.5 Cross Frame Consistent Detailing ......................................................................................................................... 132
5.5.1
5.5.2
5.6 Solid Plate Diaphragms........................................................................................................................................... 144
5.6.1
5.6.2
5.7 Temperature Change .............................................................................................................................................. 158
5.7.1
5.7.2
6
CONCLUSIONS ............................................................................................................................................................. 177
6.1 Project Overview ..................................................................................................................................................... 177
6.2 Parametric Study Findings ..................................................................................................................................... 178
7
REFERENCES ............................................................................................................................................................... 186
iii
8
APPENDIX A: LITERATURE SEARCH REPORT .................................................................................................. 188
TABLE OF CONTENTS ............................................................................................................................................... 190
8.1 INTRODUCTION ................................................................................................................................................... 191
8.2 GENERAL DESIGN GUIDELINE AND LITERATURE SEARCH ................................................................. 191
8.2.1
General Design Provisions and Guidelines ................................................................................................. 192
8.2.2
Updated Literature Search........................................................................................................................... 193
8.2.3
Summary ..................................................................................................................................................... 196
8.3 PROPOSED DESIGN FOR CONSTRUCTION DOCUMENT TEMPLATE .................................................. 196
8.4 REFERENCES ........................................................................................................................................................ 197
9
APPENDIX B: FINAL INTERIM REPORT: NUMERICAL MODELING ............................................................ 201
Table of Contents............................................................................................................................................................ 203
9.1 Introduction ............................................................................................................................................................. 204
9.2 Structure Descriptions ............................................................................................................................................ 204
9.2.1
Structure #207 ............................................................................................................................................. 204
9.2.2
Structure #7A .............................................................................................................................................. 205
9.2.3
Missing Ramps Bridge ................................................................................................................................ 206
9.2.4
Structure #28 ............................................................................................................................................... 215
9.3 Modeling Techniques .............................................................................................................................................. 216
9.3.1
Model Construction..................................................................................................................................... 216
9.3.2
Analysis Techniques ................................................................................................................................... 218
9.4 Comparisons and Modifications ............................................................................................................................ 219
9.4.1
Comparisons – Structure #207 .................................................................................................................... 219
9.4.2
Modifications .............................................................................................................................................. 237
9.5 Additional Evaluations............................................................................................................................................ 242
9.5.1
Evaluations – Curved Bridges ..................................................................................................................... 242
9.5.2
Evaluations – Skewed Bridge ..................................................................................................................... 250
9.6 Conclusions .............................................................................................................................................................. 253
9.7 References ................................................................................................................................................................ 254
Appendix A – Representative Moment and Stress Calculations ................................................................................ 255
10 APPENDIX C: PARAMETRIC STUDY BRIDGES .................................................................................................. 258
APPENDIX C-1: Curved Bridge Drawings ................................................................................................................. 259
APPENDIX C-2: Skewed Bridge Drawings ................................................................................................................. 272
11 APPENDIX D: PENNDOT DOCUMENTS ................................................................................................................ 284
iv
List of Figures
Figure 1. Structure #207 Instrument Locations. ..............................................................................3
Figure 2. Structure #314 Instrument Locations. ..............................................................................4
Figure 3. Structure #207 Section A-A Strain Variation...................................................................5
Figure 4. Structure #207 Section B-B Strain Variation. ..................................................................5
Figure 5. Structure #207 Section C-C Strain Variation. .................................................................6
Figure 6. Structure #314 Section B-B Strain Variation. .................................................................6
Figure 7. Structure #314 Section C-C Strain Variation. ..................................................................7
Figure 8. Structure #314 Section D-D Strain Variation...................................................................7
Figure 9. Structure #314 Section E-E Strain Variation....................................................................8
Figure 10. Structure #314 Section F-F Strain Variation. .................................................................8
Figure 11. Structure #314 Section G-G Strain Variation.................................................................9
Figure 12. Structure #314 Section H-H Strain Variation.................................................................9
Figure 13. Curved Bridge Statistics, Radius of Curvature. ............................................................11
Figure 14. Curved Bridge Statistics, Span Number. ......................................................................12
Figure 15. Curved Bridge Statistics, Number of Girders...............................................................12
Figure 16. Curved Bridge Statistics, Span Length. ........................................................................13
Figure 17. Curved Bridge Statistics, Girder Spacing. ....................................................................13
Figure 18. Curved Bridge Statistics, Cross-Frame Spacing. .........................................................14
Figure 19. Skewed Bridge Statistics, Skew Angle. .......................................................................15
Figure 20. Skewed Bridges Statistics, Span Number. ...................................................................15
Figure 21. Skewed Bridges Statistics, Number of Girders. ...........................................................16
Figure 22. Skewed Bridges Statistics, Span Length. .....................................................................16
Figure 23. Skewed Bridges Statistics, Girder Spacing. .................................................................17
Figure 24. Skewed Bridges Statistics, Cross-Frame Spacing. .......................................................17
Figure 25. 3-D Modeling in SAP2000 of a Representative Curved Bridge. .................................19
Figure 26. 3-D Modeling in SAP2000 of a Representative Skewed Bridge. ................................19
Figure 27. Girder Boundary Conditions. .......................................................................................20
Figure 28. Models for Initial Curved Bridge Study. ......................................................................22
Figure 29. Simplified Framing Plan, Single-Span, 4-Girder, 305-m (1000-ft) Radius
Bridge. ............................................................................................................................................24
Figure 30. Simplified Framing Plan, Single-Span, 4-Girder, 91-m (300-ft) Radius Bridge,
R/L = 13.33. ...................................................................................................................................24
Figure 31. Simplified Framing Plan, Balanced Two-Span, 4-Girder, 305-m (1000-ft)
Radius Bridge.................................................................................................................................25
Figure 32. Simplified Framing Plan, Balanced Two-Span, 4-Girder, 91-m (300-ft) Radius
Bridge, R/L = 13.33. ......................................................................................................................25
Figure 33. Simplified Framing Plan, Unbalanced Two-Span, 4-Girder, 305-m (1000-ft)
Radius Bridge.................................................................................................................................27
Figure 34. Simplified Framing Plan, Unbalanced Two-Span, 4-Girder, 91-m (300-ft)
Radius Bridge, R/L = 13.33. ..........................................................................................................28
Figure 35. Stage 1 of Construction for Paired Girder (Inner Girders Placed First) Erection
of Single-Span, 4-Girder, 305-m (1000-ft) Radius Bridge. ...........................................................28
v
Figure 37. Stage 1 of Construction for Single Girder (Inner Girder Placed First) Erection
Figure 42. Stage 2 of Construction for Paired Girder (Inner Girders Placed First)
Erection of Single-Span, 5-Girder, 305-m (1000-ft) Radius Bridge. ............................................30
of Two Equal Span, 4-Girder, 305-m (1000-ft) Radius Bridge. ....................................................32
Figure 53. Slab Pour Sequence for Two Equal Span, 4- and 5-Girder, 305-m (1000-ft)
Radius Bridges. ..............................................................................................................................33
of Two Unequal Spans, 4-Girder, 305-m (1000-ft) Radius Bridge. ..............................................33
Figure 58. Slab Pour Sequence for Two Unequal Spans, 4- and 5-Girder, 305-m (1000-ft)
Radius Bridge.................................................................................................................................34
Figure 59. Structure 7A Framing Plan. ..........................................................................................35
Figure 60. Plan View Detailing Deflection Directions. .................................................................36
vi
Figure 61. Girder 1 Bottom Flange Vertical Deflections for Single-Span, 4-Girder, 305m
(1000 ft.) Radius Bridge. ...............................................................................................................37
Figure 62. Girder 1 Bottom Flange Radial Deflections for Single-Span, 4-Girder, 305m
(1000 ft.) Radius Bridge. ...............................................................................................................38
(1000 ft.) Radius Bridge. ...............................................................................................................39
Figure 64. Girder 4 Bottom Flange Vertical Deflections for Single-Span, 4-Girder, 305m
(1000 ft.) Radius Bridge. ...............................................................................................................40
(1000 ft.) Radius Bridge. ...............................................................................................................41
Figure 66. Girder 4 Bottom Flange Tangential Deflections for Single-Span, 4-Girder,
305m (1000 ft.) Radius Bridge. .....................................................................................................42
Figure 67. Construction Method Influence on Fitted Mean Vertical Deflections. ........................50
Figure 68. Construction Method Influence on Fitted Mean Radial Deflections............................52
Figure 69. Fitted Means Radial Deflection Comparison for the Interaction of
Construction Method and R/L Ratio. .............................................................................................53
Figure 70. Construction Method Influence on Fitted Mean Tangential Deflections. ....................55
Figure 71. Fitted Mean Tangential Deflection Comparison for the Interaction of
Construction Method and Span Type. ...........................................................................................56
Figure 72. Radial Deformations, Single Girder Erection. First Erected Girder. ...........................61
Figure 73. Vertical Deformations, Single Girder Erection. First Erected Girder. .........................61
Figure 74. Radial Deformations, Single Girder Erection. Last Erected Girder. ............................62
Figure 75. Vertical Deformations, Single Girder Erection. Last Erected Girder. .........................62
Figure 76. Radial Deformations, Paired Girder Erection. First Erected Girder. ...........................63
Figure 77. Vertical Deformations, Paired Girder Erection. First Erected Girder. .........................64
Figure 78. Radial Deformations, Paired Girder Erection. Last Erected Girder. ............................64
Figure 79. Vertical Deformations, Paired Girder Erection. Last Erected Girder. .........................65
Figure 80: Simplified Framing Plan, Bridge C3: Two-Span, 4-Girder, 91.4 m Radius,
R/L=13.3 ........................................................................................................................................68
R/L=28.9 ........................................................................................................................................68
R/L=44.4 ........................................................................................................................................68
Figure 83: Simplified Framing Plan, Bridge C10: Balanced, Three-Span, 4-Girder, 91.4
Radius, R/L=13.3 ...........................................................................................................................69
Figure 84: Simplified Framing Plan, Bridge C11: Unbalanced, Three-Span, 4-Girder,
91.4 Radius, R/L=13.3 ...................................................................................................................69
Figure 85: Stage 1 to Stage 6 for Paired-Girder (inner girders placed first) Erection of
Two-Equal-Span, 4-Girder bridges. ...............................................................................................70
Figure 86: Stage 1 to Stage 6 for Paired-Girder (outer girders placed first) Erection of
Two-Equal-Span, 4- Girder Bridges. .............................................................................................71
Figure 87: Stage 1 to Stage 12 of Construction for Single-Girder (inner girder placed
first) Erection of Two-Equal-Span, 4-Girder Bridges. ..................................................................72
Figure 88: Stage 1 to Stage 12 of Construction for Single-Girder (outer girder placed
first) Erection of Two-Equal-Span, 4-Girder Bridges. ..................................................................73
Figure 89: Stage 1 to Stage 14 of Construction for ―Drop-In‖ Erection of Bridge C-10. .............75
vii
Figure 90: Stage 1 to Stage 8 of Construction for ―Drop- .............................................................75
Figure 91: Ratio of Maximum Vertical Deflections for Bridge C3. ..............................................76
Figure 94: Ratio of Maximum Vertical Deflections for Bridge C10. ............................................78
Figure 96: Ratio of Maximum Radial Deflections for Bridge C3. ................................................79
Figure 99: Ratio of Maximum Radial Deflections for Bridge C10. ..............................................80
Figure 100: Ratio of Maximum Radial Deflections for Bridge C11. ............................................81
Figure 101: Simplified Framing Plan of Bridge S2: Single-Span, 50 Skew. ...............................82
Figure 102: Simplified Framing Plan of Bridge S6: Two-Span, 50 Skew. ..................................82
Figure 103: Simplified Framing Plan of Bridge S8: Two-Span , 70 Skew. .................................82
Figure 104: Simplified Framing Plan of Bridge S9: Balanced, Three-Span, 50 Skew. ...............82
Figure 105: Simplified Framing Plan of Bridge S10: Unbalanced, Three-Span, 50 Skew..........82
Figure 106: Stage 1 to Stage 6 of Construction for Paired-Girder Erection of Two EqualSpan, 4-Girder Bridges. .................................................................................................................83
Figure 107: Stage 1 to Stage 12 of Construction for Single-Girder Erection of TwoEqual-Spans, 4-Girder Bridges. .....................................................................................................84
Figure 108: Stage 1 to Stage 10 of Construction for ―Drop-In‖ Erection of Bridge S-9 and
Bridge S-10. ...................................................................................................................................85
Figure 109: Ratio of Maximum Vertical Deflections for Bridge S2, Single-Span, 50
Skew...............................................................................................................................................86
Figure 110: Ratio of Maximum Vertical Deflections for Bridge S6, Two-Span, 50 Skew. ........86
Figure 111: Ratio of Maximum Vertical Deflections for Bridge S8, Two-Span, 70 Skew. ........87
Figure 112: Ratio of Maximum Vertical Deflections For Bridge S9, Balanced ThreeSpan, 50 Skew. .............................................................................................................................87
Figure 113: Ratio of Maximum Vertical Deflections for Bridge S10, Unbalanced ThreeSpan, 50 Skew ..............................................................................................................................87
Figure 114: Girder Web Out-Of-Plumbness Information. .............................................................89
Figure 125: Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge C1. ..................96
Figure 126: Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge C3. ..................96
Figure 127. Ratio of Maximum Vertical Deflections for Bridge C3, 6% Out-ofPlumbness. .....................................................................................................................................97
Figure 128. Ratio of Maximum Radial Deflections for Bridge C3, 6% Out-of-Plumbness. .........97
viii
Figure 129 Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge C3, 6%
Out-of-Plumbness. .........................................................................................................................98
Figure 130: Ratio of Maximum Vertical Deflections for Bridge S2. ............................................98
Figure 133: Ratio of Maximum Vertical Deflections for Bridge S7. ..........................................100
Figure 135: Ratio of Maximum Radial Deflections for Bridge S2. .............................................101
Figure 138: Ratio of Maximum Radial Reflections for Bridge S7. .............................................102
Figure 140: Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge S5. ................103
Figure 142: Ratio of Maximum Vertical Deflections for Bridge S6, 6% Out-ofPlumbness.. ..................................................................................................................................104
Figure 143: Ratio of Maximum Lateral Deformations for Bridge S6, 6% Out-ofPlumbness. ...................................................................................................................................105
Figure 144: Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge S6, 6%
Out-of-Plumbness. .......................................................................................................................105
Figure 145: Shoring Conditions for Two-Span Bridges. .............................................................109
Figure 146: Shoring Conditions for Bridge C10..........................................................................110
Figure 149: Shoring Conditions for Single-Span Skewed Bridge. ..............................................113
Figure 150: Shoring Conditions for Two-Span Skewed Bridges. ...............................................113
Figure 151: Shoring Conditions for Three-Span Skewed Bridges. .............................................114
Figure 152: Ratio of Maximum Vertical Deflections for Bridge C1. ..........................................115
Figure 157: Ratio of Maximum Vertical Deflections for Bridge C10. ........................................117
Figure 163: Ratio of Maximum Von Mises Stresses for Bridge C3. ...........................................120
Figure 166: Ratio of Maximum Von Mises Stress for Bridge C3 with Various Settlement
Conditions. ...................................................................................................................................122
ix
Figure 173: Ratio of Maximum Vertical Deflections for Bridge S10. ........................................126
Figure 174: Ratio of Maximum Vertical Deflections for Bridge S11. ........................................126
Figure 175: Ratio of Maximum Vertical Deflections for Three Shoring Conditions in
Bridge S5. ....................................................................................................................................127
Bridge S-6. ...................................................................................................................................127
Bridge S7. ....................................................................................................................................128
Bridge S8. ....................................................................................................................................128
Figure 179: Ratio of Maximum Von Mises Stresses for Bridge S5. ...........................................129
Figure 183: Ratio of Maximum Von Mises Stress for Bridge S6 with Various Settlement
Conditions. ...................................................................................................................................131
Figure 184: Cross Frame Geometries for Inconsistent Detailing. ...............................................132
Figure 188: Ratio of Maximum Radial Deflections for Bridge C11. ..........................................136
Figure 190: Ratio of Maximum Von Mises Stresses for Bridge C11. .........................................137
x
Figure 211: Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge C1. ................150
Figure 219: Ratio of Maximum Lateral Deflections for Bridge S5. ............................................154
Figure 223: Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge S5 .................156
Figure 232. Plan Views, Missing Ramps Bridge. ........................................................................209
Figure 233. Typical Cross-Sections, Missing Ramps Bridge. .....................................................209
Figure 234. Unit 1, Girder Erection Details, Missing Ramps Bridge. .........................................211
Figure 235. Unit 2, Girder Erection Details, Missing Ramps Bridge. .........................................212
Figure 236. Unit 1 Deck Placement Sequence, Missing Ramps Bridge. .....................................214
Figure 237. Unit 2 Deck Placement Sequence, Missing Ramps Bridge. .....................................215
Figure 238. Representative B31 Flange Element. .......................................................................218
Figure 239. Typical Cross-Section, Shell and Beam Model Construction. .................................218
Figure 240. Structure #207 Construction Stage. ..........................................................................220
Figure 241. Structure #207 Instrumented Sections. .....................................................................221
Figure 242. Stage 6 Vertical Bending Moment Distribution for Section B-B, Structure
#207..............................................................................................................................................222
Figure 243. Stage 8b Vertical Bending Moment Distribution for Section B-B, Structure
#207..............................................................................................................................................223
Figure 244. Stage 5 Vertical Bending Moment Distribution for Section C-C, Structure
#207..............................................................................................................................................224
Figure 245. Stage 6 Vertical Bending Moment Distribution for Section C-C, Structure
#207..............................................................................................................................................225
Figure 246. Stage 8b Vertical Bending Moment Distribution for Section C-C, Structure
#207..............................................................................................................................................226
Figure 247. Stage 8b Lateral Bending Moment Distribution for Section B-B, Structure
#207..............................................................................................................................................227
Figure 248. Stage 8b Lateral Bending Moment Distribution for Section C-C, Structure
#207..............................................................................................................................................228
Figure 249. Deck Placement Sequence, Structure #207. .............................................................229
xi
Figure 250. Stress Comparisons for Girder 1 Bottom Flange Section A-A, Structure #207. ......230
Figure 251. Stress Comparisons for Girder 1 Bottom Flange Section B-B, Structure #207. ......231
Figure 252. Stress Comparisons for Girder 3 Top Flange Section B-B, Structure #207. ............232
Figure 253. Change in Girder 1 Vertical Displacements between Erection of Entire Steel
Superstructure and Placement of the Slab, Structure #207. .........................................................233
Figure 258. Vertical Deflection Comparison, Concentrated Load at Midspan, Heins and
Spates (1968) Tests. .....................................................................................................................238
Figure 259. Rotation Comparison, Concentrated Load at Three-Tenths Point, Heins and
Spates (1968) Tests. .....................................................................................................................239
Figure 260. Vertical Deflection Comparisons, Original and Modified Slab Placement
Techniques, Structure #207. ........................................................................................................240
Figure 261. Radial Deflection Comparisons, Original and Modified Slab Placement
Techniques, Structure #207. ........................................................................................................241
Figure 262. Stage 8b Vertical Bending Moment Distribution for Section C-C, Including
Flange Offset, Structure #207. .....................................................................................................242
Figure 263. ABAQUS Model, Structure #7A. .............................................................................243
Figure 264. G1Vertical Displacement Changes During Realignment of Spans 4 and 5,
Structure #7A. ..............................................................................................................................244
Structure #7A. ..............................................................................................................................244
Structure #7A. ..............................................................................................................................245
Structure #7A. ..............................................................................................................................246
Figure 268. G5 Vertical Displacement Changes During Realignment of Spans 4 and 5,
Structure #7A. ..............................................................................................................................246
Figure 269. ABAQUS Model, Missing Ramps Bridge. ..............................................................247
Figure 270. ABAQUS Model Detailing Cross-Frame Labels, Missing Ramps Bridge. .............247
Figure 271. Girder Radial Translations at Web-Flange Junction, Cross-Frame 10,
Completion of Unit 1 Girder Erection, Missing Ramps Bridge. .................................................248
Figure 274. ABAQUS Model, Structure #28...............................................................................251
Figure 275. Girder Lateral Displacements, Completion of Deck Pour, Structure #28. ...............251
Figure 276. G1 Vertical Displacements, Completion of Deck Pour, Structure #28. ...................252
Figure 277. G7 Vertical Displacements, Completion of Deck Pour, Structure #28. ...................252
xii
List of Tables
Table 1. Preliminary Curved Steel Bridge Global Information. ....................................................14
Table 2. Preliminary Skewed Steel Bridge Global Information ....................................................18
Table 3. Final Curved Bridge Global Information.........................................................................21
Table 4. Final Skewed Bridge Global Information. .......................................................................21
Table 5. Single-Span Bridge Proportions. .....................................................................................23
Table 6. Balanced, Two-Span Bridge Proportions. ......................................................................24
Table 7. Unbalanced, Two-Span Bridge Proportions ....................................................................27
Table 8. Independent ANOVA Variables. .....................................................................................42
Table 9. R/L Ratios Used in ANOVA Analysis. ...........................................................................42
Table 10. ANOVA Results for Initial Vertical Deflection Statistical Model. ..............................46
Table 11. ANOVA Results for Modified Vertical Deflection Statistical Model. ..........................47
Table 12. ANOVA Results for Final Vertical Deflection Statistical Model. ................................49
Table 13. ANOVA Results for Final Radial Deflection Model.....................................................51
Table 14. ANOVA Results for Final Tangential Deflection Model. .............................................55
Table 15. Recommended Construction Methods for Single-Span Bridges, Initial Studies. ..........57
Table 16. Construction Methods Not Recommended for Single-Span Bridges, Initial
Studies. ...........................................................................................................................................57
Table 17. Recommended Construction Methods for Balanced Two-Span Bridges, Initial
Studies. ...........................................................................................................................................58
Table 18. Construction Methods Not Recommended for Balanced Two-Span Bridges,
Initial Studies. ................................................................................................................................58
Table 19. Recommended Construction Methods for Unbalanced Two-Span Bridges,
Initial Studies. ................................................................................................................................58
Table 20. Construction Methods Not Recommended for Unbalanced Two-Span Bridges,
Initial Studies. ................................................................................................................................59
Table 21: Selected Curved Bridge Erection Study Bridge Information ........................................68
Table 22: Information From Representative Bridges. ...................................................................81
Table 23 Selected Curved Bridge Web-Plumbness Study Information.........................................90
Table 24 Selected Skewed Bridge Web-Plumbness Study Information. .......................................90
Table 25: Selected Curved Bridge Temporary Shoring Study Information. ...............................108
Table 26: Selected Settlement Conditions. ..................................................................................112
Table 27: Selected Skewed Bridge Temporary Shoring Study Information. ..............................112
Table 28: Selected Inconsistent Detailing Study Bridge Information .........................................134
Table 29: Selected Skewed Bridge Consistent Detailing Study Bridge Information ..................134
Table 30: Selected Curved Bridge Diaphragm Study Bridge Information. .................................144
Table 31: Selected Curved Bridge Diaphragm Study Bridge Information. .................................145
Table 32: Selected Curved Bridge Temperature Change Study Information. .............................159
Table 33: Selected Skewed Bridge Temperature Change Study Information. ............................160
Table 34. Summary Table. ...........................................................................................................181
Table 35. Girder Lengths, Radii, and Field Splice Locations (see Figure 1). .............................207
Table 36. Structure #207 Examined Construction Stages. ..........................................................219
xiii
1
INTRODUCTION
This report summarizes research activities and findings in association with Work Order 009, ―Guidelines
for Analyzing Curved and Skewed Bridges and Designing Them for Construction.‖ The report is intended
to serve as a compendium of activities related to the project and, as such, follows the final Work Order 009
scope document with respect to organization.
As discussed in the scope document, the objectives of the project were to: (1) continue to develop and
maintain remote acquisition capabilities (via cell phone) for instruments on two structures in the I-99
corridor – #207, a horizontally curved, steel, I-girder bridge, and #314, a skewed, pre-stressed, concrete
bridge; (2) develop, examine, and reduce data produced from these structures as needed; (3) continue
examination of numerical model accuracy for curved and skewed, steel, I-girder bridges and select
appropriate model types and software; (4) extend numerical studies to examine prevalent issues affecting
curved and skewed steel, I-girder bridge behavior during construction (detailed under tasks below and
within the enclosed draft guidelines document); and (5) develop relevant guidelines for curved and skewed,
steel, I-girder bridges during construction.
The project was divided into seven tasks, listed below:
1.
2.
3.
4.
5.
6.
7.
Updated Literature Search
Data Acquisition System Maintenance
Numerical Modeling
Parametric Studies
Draft Final Report
Final Report
Invoice Submission
Tasks that are relevant to this report are numbers 1 through 4. Each will be discussed in the sections that
follow.
2
UPDATED LITERATURE SEARCH
As discussed in the project scope, an updated literature search was completed that summarized recent
publications relevant to topics of interest on curved and skewed bridges. These topics included items
related to construction response, numerical modeling, information on the development of any design for
construction documents, and information on areas more specific to the parametric studies that were
completed. More specific topics that were researched included: the effects of bearing type and restraint on
curved and skewed, steel, I-girder behavior; the effects of temperature on curved and skewed, steel, I-girder
behavior during construction; the effects of web-plumbness and twist imperfections on curved and skewed,
steel, I-girder behavior; the effects of connection detailing on curved and skewed, steel, I-girder behavior;
and the influence of solid plate diaphragms and cross frames on curved and skewed, steel, I-girder behavior.
An updated literature search report summarizing the information obtained was submitted to PennDOT in
August 2008 (Linzell et al. 2008). The main body of that report is included as Appendix A of this
document.
3
DATA ACQUISITION SYSTEM MAINTENANCE
As discussed previously and in the project scope, bridge data acquisition and power supply systems were
maintained and monitored throughout the project on Structures 207 and 314. Data recorded during the
current project encompassed monitoring the effects of temperature cycles on the structures.
1
3.1
Instrumentation Summary
Summaries of the structures, instruments used, and their quantities and locations can be found in earlier
submittals to PennDOT (Linzell et al. 2003; Hiltunen et al. 2004; Linzell et al. 2006).To assist with
discussions of results that were obtained, short summaries of the instrumentation on each of the structures
are provided in the sections that follow.
3.1.1
Structure #207
A combination of vibrating wire strain gages and vibrating wire tiltmeters were used to monitor
superstructure response. Instruments were placed at various locations on the girders and cross frames as
shown in Figure 1. Data were acquired at specified increments throughout the construction process and
after the bridge was placed into service.
2
3
G5
G4
G3
G2
G1
VARIES
G1 - 48'-4"
G5 - 46'-10" A
A
WEST ABUTMENT
F.S. 1
B
B
N
F.S. 3
S
VARIE 5"
'6
9
1
G
'-11"
G5 - 92
C
C
Figure 1. Structure #207 Instrument Locations.
D
EAST ABUTMENT
D
GIRDER VW STRAIN GAUGE - SCHEME A (DETAIL 1)
GIRDER VW STRAIN GAUGE - SCHEME B (DETAIL 2)
CROSS FRAME VW STRAIN GAGE (DETAIL 3 & 4)
VW TILTMETER (DETAIL 5)
F.S. 4
BRIDGE S-207 - INSTRUMENT LOCATION PLAN
VARIES
G1 - 56'-11"
G5 - 56'-9"
F.S. 2
PIER
3.1.2
Structure #314
Similar to Structure #207, a combination of vibrating wire strain gages and tiltmeters were used to measure
superstructure response for Structure #314. Strain gages were mounted to the girders and diaphragms, and
tiltmeters were mounted to the girders with instruments being placed at various times during construction.
Figure 2 details the instrumented sections on the bridge. Data were acquired at specified increments
throughout the construction process and after the bridge was placed into service.
H
RT
O G
N RIN
L
C EA
A
B
B
GIRDER 1
G
N
RI G
A RIN
BE EA
L
B
C L
C
D
C
E
F
G GIRDER
7
H
TH
U G
SO RIN
CL EA
B
GIRDER 8
GIRDER 2
GIRDER 9
GIRDER 3
GIRDER 10
GIRDER 4
GIRDER 11
GIRDER 5
GIRDER 12
GIRDER 6
B
A
C
E
D
F
G
R
IE
CL
H
P
Figure 2. Structure #314 Instrument Locations.
3.2
Data Collection
As stated earlier, data collection summarized herein involved monitoring in-service bridge behavior. Given
the instruments and data acquisition systems that were used, this largely consisted of tracking responses
due to temperature changes. Data that deviated from what was anticipated or historically recorded for a
given instrument could be identified. Data were downloaded from the acquisition systems at both structures
approximately monthly throughout the duration of the project. Data readings that were downloaded were
recorded by the acquisition systems at one-hour intervals. The time period over which recorded data is
provided is between August 2008 and April 2010.
3.3
Data Reduction and Results
Downloaded data were merged with values for each instrument from previous downloads. Since weigh-inmotion systems were not included with the instrumentation suite for the bridges, data were plotted with
respect to time for each instrument. Representative plots for each structure are provided in the sections that
follow.
3.3.1
Structure #207
Representative changes in strain during the data recording period are provided at the instrumented locations
discussed above. Data are plotted as response verses time in days and provide a snapshot and some insight
related to gross changes in bridge behavior. Observed strains did not vary by more than approximately 200
during the representative period. No obvious trends were noted from the data.
4
207-G3 AA
1,000.0
750.0
500.0
Strain (x10-6)
250.0
G3 AA TF N
0.0
0
50
100
150
200
250
300
350
400
450
-250.0
G3 AA BF N
G3 AA BF S
-500.0
-750.0
-1,000.0
Days
Figure 3. Structure #207 Section A-A Strain Variation.
207-G2 BB
1,000.0
750.0
500.0
Strain (x10-6)
250.0
G2 BB TF N
0.0
0
50
100
150
200
250
300
350
-250.0
400
450
G2 BB BF N
G2 BB BF S
-500.0
-750.0
-1,000.0
Days
Figure 4. Structure #207 Section B-B Strain Variation.
5
207-G1 CC
1,000.0
750.0
500.0
Strain (x10-6)
250.0
G1 CC TF S
0.0
G1 CC BF N
0
50
100
150
200
250
300
350
400
450
G1 CC BF S
-250.0
-500.0
-750.0
-1,000.0
Days
Figure 5. Structure #207 Section C-C Strain Variation.
3.3.2
Structure #314
Data are again plotted verses time in days and provide a snapshot and some insight related to gross changes
in bridge behavior. Observed strains did not vary by more than approximately 30
during the
representative period. Again, no obvious trends were noted from the data.
314-G6 BB
100.0
75.0
50.0
Strain (x10-6)
25.0
G6 BB TF N
0.0
0
50
100
150
200
250
-25.0
300
350
G6 BB BF N
G6 BB BF S
-50.0
-75.0
-100.0
Days
Figure 6. Structure #314 Section B-B Strain Variation.
6
314-G1 CC
100.0
75.0
50.0
Strain (x10-6)
25.0
G1 CC TF S
0.0
0
50
100
150
200
250
300
350
-25.0
G1 CC BF N
G1 CC BF S
-50.0
-75.0
-100.0
Days
Figure 7. Structure #314 Section C-C Strain Variation.
314-G1 DD
100.0
75.0
50.0
Strain (x10-6)
25.0
G1 DD TF S
0.0
0
50
100
150
200
250
300
350
G1 DD BF S
-25.0
-50.0
-75.0
-100.0
Days
Figure 8. Structure #314 Section D-D Strain Variation.
7
314-G12 EE
100.0
75.0
50.0
Strain (x10-6)
25.0
G12 EE BF N
0.0
0
50
100
150
200
250
300
350
G12 EE BF S
-25.0
-50.0
-75.0
-100.0
Days
Figure 9. Structure #314 Section E-E Strain Variation.
314-G9 FF
100.0
75.0
50.0
Strain (x10-6)
25.0
G9 FF TF S
0.0
0
50
100
150
200
250
-25.0
300
350
G9 FF BF N
G9 FF BF S
-50.0
-75.0
-100.0
Days
Figure 10. Structure #314 Section F-F Strain Variation.
8
314-G9 GG
100.0
75.0
50.0
Strain (x10-6)
25.0
G9 GG TF S
0.0
0
50
100
150
200
250
300
350
-25.0
G9 GG BF N
G9 GG BF S
-50.0
-75.0
-100.0
Days
Figure 11. Structure #314 Section G-G Strain Variation.
314-G7 HH
100.0
75.0
50.0
Strain (x10-6)
25.0
0.0
0
50
100
150
200
250
300
350
G7 HH NA ST
-25.0
-50.0
-75.0
-100.0
Days
Figure 12. Structure #314 Section H-H Strain Variation.
4
NUMERICAL MODELING
As discussed in the scope, work was completed that established appropriate model types for application in
the parametric studies that are outlined and discussed in sections that follow. Models of differing
complexities were developed and compared to field data from the structures discussed earlier. Some
additional structures and, via these comparisons, a recommended modeling scheme were developed for the
parametric studies.
9
A numerical modeling report, summarizing work that was done and decisions that were made to establish
the modeling procedure outlined below, was initially submitted to PennDOT in August 2008 (Linzell et al.
2008). The main body of that report is included as Appendix B of this document.
4.1
Parametric Study Modeling Procedure
As discussed in the previously submitted numerical modeling report (Linzell et al. 2008), models that were
recommended for the parametric studies consisted of shell elements for the girder webs and bridge decks
and beam elements for the girder flanges and cross frame members. The shell elements used for the girder
webs and the bridge deck were ABAQUS S4R elements having 1:1 and 1:2 aspect ratios, respectively. The
beam elements used for flanges were ABAQUS B31 beam elements. The report also suggested that using a
first-order deformation analysis provides an effective method to predict deflections, including rotations, for
a bridge with severe horizontal curvature. Therefore, analyses were performed using a first-order
deformation approach.
The analysis procedure for each parametric study discussed in the sections that follow consisted of
sequential analysis of curved and skewed bridges that incorporated various parameters examined in this
study. The sequential analyses were performed by creating multiple steps involving analyses of each
construction stage. During each step, the program analyzed the structure having the given structural
components, loads, and boundary conditions based on that construction stage. The program completed the
analysis when all girders were erected and the concrete deck was placed. Results from the sequential
analyses were used to examine the effects of those parameters on bridge constructability. Details on how
each parameter was incorporated in the finite element model are described in the following chapter.
5
PARAMETRIC STUDIES
The majority of the work discussed in the current report relates to this task, which encompasses a number
of aspects as discussed in the project scope. In a general sense, the parametric studies that were performed
examined the influence of construction methodology and sequencing on the resulting stresses and
deformations in both curved and skewed, steel, I-girder bridges.
More specifically, the parametric studies examined the effects of a specific number of variables involved in
curved, steel, I-girder bridge construction on the response of a group of representative bridge structures,
which encompassed a specific range of geometries. Variables that were examined as the studied structures
were numerically ―constructed‖ included: (1) web-plumbness; (2) temporary shoring placement and
settlement effects; (3) cross-frame consistent detailing (i.e., applying the work by Chavel and Earls to other
bridge geometries); (4) girder and cross-frame erection sequencing along the span and with respect to
girder radius and the effects of ―drop-in‖ erection; (5) solid plate diaphragms verses cross frames; and (6)
global temperature change during placement of the deck.
5.1
Parametric Structure Selection and Design
Background information related to how the representative structures were selected and designed is provided
prior to discussing any information related to the parametric studies and their subsequent results. The
discussion includes information related to: statistical studies completed to assist with significant curved and
skewed bridge global design variable identification and selection; down-select decisions used to establish
the final representative curved and skewed bridge designs used for the parametric studies; and specific
information related to how final proportions were obtained for the representative bridges that were
analyzed.
5.1.1
Initial Bridge Parameter Identification and Final Structure Down Selection
The identification and selection of significant parameters and parameter ranges used to initiate design of the
representative curved and skewed bridges for the parametric studies began with an inventory survey of a
10
large set of bridge design plans from Maryland, New York, and Pennsylvania. A total of 355 bridges,
consisting of horizontally curved, steel, I-girder structures with and without skew, were included in the
inventory study. Of these 355 bridges, 129 had no skew (36% of the total) and 226 were a combination of
skewed and curved steel I-girder bridges (64% of the total).
5.1.1.1
Curved Bridges Statistics
The statistical study completed herein focused on curved bridges with no skew and having a single center
of curvature. This initial examination assisted with establishing the initial design parameters for curved
bridges and provided some input for the initial design parameters for skewed bridges. Of the 129 curved
bridges that had no skew, 102 were constructed using a single horizontal curve. These 102 bridges were
investigated in more detail to identify statistically significant values of selected global geometric
parameters. Curved bridge parameters that were investigated included the radius of curvature, span and
girder numbers, span length, and girder and cross-frame spacing. Figure 13 through Figure 18 show the
resulting frequency distributions for each of these parameters from the bridge sample.
Figure 13. Curved Bridge Statistics, Radius of Curvature.
11
Figure 14. Curved Bridge Statistics, Span Number.
Figure 15. Curved Bridge Statistics, Number of Girders.
12
Figure 16. Curved Bridge Statistics, Span Length.
Figure 17. Curved Bridge Statistics, Girder Spacing.
13
Figure 18. Curved Bridge Statistics, Cross-Frame Spacing.
The study of the bridge statistics in conjunction with examining the current design specifications was a
starting point for developing an initial, proposed set of curved bridges that would be examined
parametrically. Twelve horizontally curved steel bridges, shown in Table 1, were preliminarily selected for
the planned parametric studies. They consisted of nine balanced, two-span structures and three balanced
and unbalanced three-span continuous structures. Preliminary analyses and designs of these structures were
completed with the aid of SAP2000, and the process that was followed is presented in subsequent sections.
Bridge No.
1
2
3
4
5
6
7
8
9
5.1.1.2
Table 1. Preliminary Curved Steel Bridge Global Information.
Radius of
Cross-Frame
GirderSpan-Length,
Number of
Curvature, ft
Spacing, ft
Spacing, ft
ft
Girder, ft
300
15
8
225
4
300
18
8
225
4
300
22.5
8
225
4
650
15
8
225
4
650
18
8
225
4
650
22.5
8
225
4
1000
15
8
225
4
1000
18
8
225
4
1000
22.5
8
225
4
Skewed Bridges Statistics
For the skewed bridges, similar statistical analyses were completed using the same inventory for the curved
bridges to identify global geometric parameters of interest. This study focused only on bridges with skewed
abutments. The parameters investigated for skewed bridges included the skew angle, span and girder
numbers, span length, and girder and cross-frame spacing. Figure 19 through Figure 24 show the resulting
frequency distributions for each of these parameters from the bridge sample.
14
Figure 19. Skewed Bridge Statistics, Skew Angle.
Figure 20. Skewed Bridges Statistics, Span Number.
15
Figure 21. Skewed Bridges Statistics, Number of Girders.
Figure 22. Skewed Bridges Statistics, Span Length.
16
Figure 23. Skewed Bridges Statistics, Girder Spacing.
Figure 24. Skewed Bridges Statistics, Cross-Frame Spacing.
Based on the statistical analysis, an initial, proposed set of skewed bridges that would be examined
parametrically was defined. Nine straight, skewed steel bridges, shown in Table 2, were preliminarily
selected for the planned parametric studies. They consisted of four single-span structures, four balanced,
two-span structures, and three balanced and unbalanced three-span continuous structures. Preliminary
analyses and designs of these structures were completed with the aid of SAP2000, and the process that was
followed to finalize the group that would be examined parametrically is presented in subsequent sections.
17
Bridge
No.
1
2
3
4
5
6
7
8
9
5.1.2
5.1.2.1
Table 2. Preliminary Skewed Steel Bridge Global Information
CrossNumber
Skew
Frame
Girderof
Angle,
Span-Length, ft
Spacing,
Spacing, ft
Spans
degree
ft
50
15
10
1
168
10
60
20
1
168
10
80
24
1
168
10
50
15
2
168-168
10
60
20
2
168-168
10
80
24
2
168-168
10
50
24
3
168-168-168
10
50
24
3
120-168-168
10
50
24
3
120-168-120
Number
of Girders
4
4
4
4
4
4
4
4
4
Structure Design and Final Proportions
Preliminary Designs
Based on the statistical analyses discussed in the previous section, initial bridge global geometries were
selected and preliminary designs were completed for both the curved and skewed bridges. Designs for the
concrete deck design and steel superstructure were completed following relevant chapters from the
AASHTO LRFD Bridge Design Specifications (AASHTO, 2007) and the PennDOT Design Manual
(PennDOT, 2007). Irrespective of structure type, all bridges were initially assumed to support two lanes of
traffic with shoulders and barriers; these assumptions resulted in an initial deck width of 38 ft. The deck
was designed assuming one-way action and acted compositely with the supporting plate girders. Its
thickness was selected to satisfy the requirements for flexure and shear strength limit states according to
AASHTO and PennDOT. The initial steel superstructures all consisted of four plate girders spaced at 10
feet that were braced using X-shape cross frames with top and bottom cords. Lower lateral bracing was also
included in the initial designs. Preliminary plate girder dimensions were selected for both curved and
skewed bridge sets following the cross-section proportion limits from relevant AASHTO and PennDOT
DM4 articles. Plate girders were assumed to be braced at the maximum allowable limits for curved plate
girders from AASHTO. Although there is no unbraced length limit for straight plate girders in the
specifications, to have an efficient and consistent design for the skewed structures that avoided lateral
torsional buckling, applicable limits from AASHTO for curved plate girders were also used for the girders
in the skewed bridges. To have an efficient cross section design, flanges were sized to be either compact or
non-compact, and webs were sized to be slender following the slenderness limits from AASHTO and
PennDOT DM4. Cross frames and horizontal bracings were initially sized for anticipated wind loads on
girder-developed surfaces, and they were initially designed assuming they acted solely as tension or
compression members. Initial sizes for all the bridge components were used to create models in SAP2000
that assisted with completion of the final designs.
5.1.2.2
Final Designs
Analyses that helped with establishing final curved and skewed bridge designs were completed using 3D
SAP2000 finite element models. In these models, the plate girders were represented using shell elements
for the top and bottom flanges and the webs. The concrete deck was also modeled using shell elements.
Cross frames were incorporated into the models using frame elements. Representative curved and skewed
bridge models are shown in Figure 25 and Figure 26, respectively. All plate girders were assumed to be
homogeneous with a yield stress of 50 ksi, and the deck concrete was assumed to have a 4 ksi compressive
strength. Single-span bridges were assumed to be simply supported, while multi-span structures had
18
pinned supports at one interior pier and roller supports everywhere else. Figure 27 details the boundary
conditions applied to the bridges.
Figure 25. 3-D Modeling in SAP2000 of a Representative Curved Bridge.
Figure 26. 3-D Modeling in SAP2000 of a Representative Skewed Bridge.
19
Figure 27. Girder Boundary Conditions.
Loading Criteria
Refined loads were then used to modify the initial plate girder and cross frames section geometries. Self
weights for bridge structural components were obtained directly from preliminary design member sizes
(DC1 in AASHTO). Dead loads due to wearing surface (DW) and other non-structural components (DC2)
were applied to the concrete deck as distributed loads. HL-93 vehicular live loads were applied to the deck
as well.
Plate Girder Flexural Design
The AASHTO and PennDOT DM4 STRI load combination was used for strength limits for flexure, with
Service II being used for service limits and 1.25*DC1 being used for constructability limits. For both
curved and skewed bridges, final section designs changed as a function of positive and negative moment
magnitudes. Splices were located at dead-load contraflexure points. Where section sizes could be modified
due to positive and negative moment magnitudes, factors such as web depths, web thicknesses, and top and
bottom flange widths were kept constant, and only top and bottom flange thicknesses changed. Following
the preliminary designs, webs were slender in all cross sections, and top and bottom flanges were either
compact or non-compact.
Plate Girder Shear Design
AASHTO and PennDOT DM4 STRI were used to check strength limits for shear, and 1.25*DC1 was used
to check constructability limits. To enhance the web shear capacity, transverse stiffeners were added to the
plate girder web. Bearing stiffeners were also included in the final design. These stiffeners were sized in
accordance with AASHTO.
Cross Frame Refined Design
According to the design specifications, cross frames are considered to be primary members in curved and
severely skewed bridges, and so their initial cross section geometries were checked again here against the
final design loads. AASHTO and PennDOT DM4 design provisions for compression and tension members
were considered again for the final design of these members. The lateral bracing were designed as
secondary members according to the specifications, and their preliminary designs were used in the final
design.
Based on design refinements and section optimization as well as input from PennDOT personnel related to
design decisions, two final sets of 12 horizontally curved and 11 skewed bridges were obtained for
examination in the parametric studies that are summarized in the following sections. The final design
20
global parameters are summarized in Table 3 and Table 4. Corresponding plans for those bridges are found
in Appendix C. These final bridge geometries were used to create numerical models in ABAQUS for the
parametric studies.
Table 3. Final Curved Bridge Global Information.
Bridge
No.
Radius of
Curvature, ft
C1
C2
C3
C4
C5
C6
C7
C8
C9
C10
C11
C12
300
300
300
650
650
650
1000
1000
1000
300
300
300
CrossFrame
Spacing, ft
15
18
22.5
15
18
22.5
15
18
22.5
22.5
22.5
22.5
GirderSpacing, ft
Number
of Spans
Span-Length, ft
Number of
Girders
2
2
2
2
2
2
2
2
2
3
3
3
225-225
225-225
225-225
225-225
225-225
225-225
225-225
225-225
225-225
225-225-225
157.5-225-225
157.5-225-157.5
4
4
4
4
4
4
4
4
4
4
4
4
10
10
10
10
10
10
10
10
10
10
10
10
Table 4. Final Skewed Bridge Global Information.
5.2
Bridge
No.
Skew
Angle,
degree
S1
S2
S3
S4
S5
S6
S7
S8
S9
S10
S11
50
50
70
70
50
50
70
70
50
50
50
CrossFrame
Spacing,
ft
15
25.7
15
25.7
15
25.7
15
25.7
25.7
25.7
25.7
GirderSpacing, ft
Number
of
Spans
10
10
10
10
10
10
10
10
10
10
10
1
1
1
1
2
2
2
2
3
3
3
Span-Length, ft
Number
of Girders
180
180
180
180
180-180
180-180
180-180
180-180
180-180-180
180-180-128.75
128.75-180-128.75
4
4
4
4
4
4
4
4
4
4
4
Girder and Cross-Frame Erection Sequencing
The effects of erection sequencing decisions on curved and skewed bridge responses are the first parametric
study item discussed here, as findings from this portion of the project influenced the other parametric
studies that were completed.
Studying the effects of erection sequencing began with the examination of their effects on curved, I-girder
structures. Prior to the final curved structure down-select for the parametric studies outlined herein,
examinations of a broader range of single- and two-span structures and of a single three-span structure were
conducted to ascertain erection performance and assist with sequencing recommendations. Discussion of
those studies, their findings, and how those findings were applied to the selected curved bridge set are
21
presented in the sections that follow. The final curved structures were selected, based on those findings, to
include significant parameters that may influence the effects of erection sequencing methods on bridge
response. Various erection sequencing scenarios used in the initial studies were applied to the final
structures to reaffirm findings from the initial studies.
For the skewed structures, final structures were selected to include variables similar to those used for
curved structures. Similar erection sequencing scenarios used for curved structures were applied to the
skewed structures to study their effects.
5.2.1
Parametric Studies - Curved
5.2.1.1
Initial Studies - Background
The initial set of studies involving single- and two-span structures examined a total of thirty bridges under
the influence of the following erection sequencing parameters: (1) varying radii; (2) single-span structures
and two-span structures with varying span ratios; (3) four- and five-girder cross sections; (4) different
erection sequencing options that including erecting single girders and girders in pairs; (5) erecting the
girders from inner to outer radius of curvature and from outer to inner radius of curvature; and (6) the
influence of temporary shoring.
As shown in Figure 28, the three centerline radii investigated for these initial studies were 91.4 m (300.0
ft), 198.1 m (650.0 ft), and 304.8 m (1000.0 ft) and were intended to represent a severely curved bridge
(91.4 m), a moderately curved bridge (304.8 m), and a curved bridge that fell in between the extremes
(198.1 m). Three different R/L values, which represent the ratio of radius of the girder to unbraced length,
were investigated for the severely curved bridge: 10, 13.33, and 20. The 198.1-m (650.0-ft) and 304.8-m
(1000.0-ft) bridges were studied for a single R/L ratio. Based on the parameters being considered and their
combinations, a total of 240 different scenarios were investigated. Finite element models of the bridges
were created and analyzed in ABAQUS/Standard using the recommended modeling procedure discussed
earlier and in previous submittals to PennDOT. The structures were analyzed for all construction stages,
including placement of the concrete deck. Figure 28 summarizes the scenarios that were analyzed for this
preliminary study.
Base Models
Two Balanced
Spans
Single Span
4 girders
R=300'
R=650'
R=
1000'
Unbraced
Length = 22.5'
(R/L=13.33)
Unbraced
Length = 18'
(R/L=16.67)
4 girders
R=300'
R=300'
Unbraced
Length = 22.5'
(R/L=13.33)
5 girders
R=650'
Unbraced
Length = 18'
(R/L=16.67)
Unbraced
Length = 15'
(R/L=20)
R=650'
Two Unbalanced
Spans
4 girders
5 girders
R=1000'
R=300'
R=650'
R=1000'
R=300'
5 girders
R=650'
R=1000'
R=300'
R=650'
R=1000'
R=1000'
Unbraced
Length = 22.5'
(R/L=13.33)
Unbraced
Length =
15'(R/L=20)
Unbraced
Length = 22.5'
(R/L=13.33)
Unbraced
Length = 18'
(R/L=16.67)
Unbraced
Length = 15'
(R/L=20)
Unbraced
Length = 18'
(R/L=16.67)
Unbraced
Length = 22.5'
(R/L=13.33)
Figure 28. Models for Initial Curved Bridge Study.
22
Unbraced
Length = 22.5'
(R/L=13.33)
Unbraced
Length = 15'
(R/L=20)
Unbraced
Length = 18'
(R/L=16.67)
Unbraced
Length = 15'
(R/L=20)
Unbraced
Length = 18'
(R/L=16.67)
Unbraced
Length = 15'
(R/L=20)
The thirty bridges shown in the figure were designed following the AASHTO Guide Specification
(AASHTO 2003) and the AASHTO LRFD Bridge Design Specifications (AASHTO 2008). Girders were
cambered vertically, but bridge superelevation and vertical curvature were not considered. Grade 50 steel
was used, and all flanges were compact and all webs transversely stiffened. A girder spacing of 2.4 m (8.0
ft) was used for all of the designs along with a 1.2-m (4.0-ft) deck overhang and a 203-mm (8-in) slab
thickness. Bearing types and locations, along with cross-frame member proportions, were kept as similar as
possible for all structures. SAP2000 grillage models were used to assist with the initial design and design
optimization for all bridges.
Resulting single-span bridges had a span length of 68.6 m (225.0 ft), and all girders were prismatic. Table 5
details the proportions of all steel superstructure components for the single-span four- and five-girder
bridge designs. Figure 29 and Figure 30 show some representative four-girder framing plans.
Bridge (No. of
Girders;
Radius, m (ft);
R/L Ratio
4, 305 (1000),
35.59
4, 198 (650),
23.13
4, 91 (300),
13.33
4, 91 (300),
16.67
4, 91 (300),
20.00
5, 305 (1000),
35.59
5, 198 (650),
23.13
5, 91 (300),
13.33
5, 91 (300),
16.67
5, 91 (300),
20.00
Table 5. Single-Span Bridge Proportions.
Transverse
Stiffener
Dimensions,
Web
Flange
mm (in) and
Dimensions,
Dimensions,
Cross Frame
Spacings, m
mm (in)
mm (in)
Members
(ft)
9.525 x 152
2794 x 22.225
63.5 x 711
L3x2.5x0.25
(0.375 x 6)
(110 x 0.875)
(2.5 x 28)
@ 4 (13)
15.875 x 216
2845 x 25.400
76.2 x 838
L3.5x2.5x0.5
(0.625 x 8.5)
(112 x 1.00)
(3.0 x 33)
@ 2.4 (8.0)
15.875 x 216
2896 x 31.750 120.65 x 1118
L6x6x0.4375
(0.625 x 8.5)
(114 x 1.250)
(4.75 x 44)
@ 2.9 (9.5)
15.875 x 216
2896 x 31.750 120.65 x 1118
L6x6x1
(0.625 x 8.5)
(114 x 1.250)
(4.75 x 44)
@ 2.9 (9.5)
15.875 x 216
2896 x 31.750 120.65 x 1118
L6x6x0.375
(0.625 x 8.5)
(114 x 1.250)
(4.75 x 44)
@ 2.9 (9.5)
12.7 x 178 (0.5
2794 x 22.225
57.15 x 711
L3x2.5x0.25
x 7)
(110 x 0.875)
(2.25 x 28)
@ 4 (13)
15.875 x
2845 x 25.400
76.2 x 838
209.55 (0.625
L3.5x2.5x0.5
(112 x 1.00)
(3.0 x 33)
x 8.25) @ 2.7
(9.0)
19.05 x 305
2896 x 31.750 120.65 x 1118
L6x6x0.4375
(0.75 x 12)
(114 x 1.250)
(4.75 x 44)
@ 2.9 (9.5)
19.05 x 305
2896 x 31.750 120.65 x 1118
L6x6x1
(0.75 x 12)
(114 x 1.250)
(4.75 x 44)
@ 2.0 (9.5)
19.05 x 305
2896 x 31.750 120.65 x 1118
L6x6x0.375
(0.75 x 12)
(114 x 1.250)
(4.75 x 44)
@ 2.0 (9.5)
23
Bearing
Stiffener
Dimensions,
mm (in)
25 x 254
(1.0 x 10)
25 x 254
(1.0 x 10)
25 x 356
(1.0 x 14)
25 x 305
(1.0 x 12)
32 x 406
(1.25 x 16)
25 x 254
(1.0 x 10)
31.75 x 254
(1.25 x 10)
35 x 356
(1.375 x 14)
35 x 356
(1.375 x 14)
35 x 356
(1.375 x 14)
Figure 29. Simplified Framing Plan, Single-Span, 4-Girder, 305-m (1000-ft) Radius Bridge.
Figure 30. Simplified Framing Plan, Single-Span, 4-Girder, 91-m (300-ft) Radius Bridge, R/L =
13.33.
Similar to the single-span bridge, the balanced, two-span bridges had span lengths of 68.6 m (225.0 ft), and
again, prismatic sections were used for the initial studies. Table 6 details proportions of all steel
superstructure components for the balanced, two-span, four- and five-girder bridge designs. Figure 31 and
Figure 32 show some representative four-girder framing plans.
Bridge (No. of
Girders;
Radius, m (ft);
R/L Ratio
4, 305 (1000),
35.59
4, 198 (650),
23.13
4, 91 (300),
13.33
4, 91 (300),
16.67
4, 91 (300),
20.00
5, 305 (1000),
35.59
5, 198 (650),
Table 6. Balanced, Two-Span Bridge Proportions.
Transverse
Stiffener
Dimensions,
Web
Flange
mm (in) and
Dimensions,
Dimensions,
Cross Frame
Spacings, m
mm (in)
mm (in)
Members
(ft)
2794 x 22.225
50.8 x 559
L3.5x2.5x0.25
9.525 x 152
(110 x 0.875)
(2 x 22)
(0.375 x 6)
@ 4 (13)
2845 x 25.4
50.8 x 610
L3.5x2.5x0.5
15.875 x 216
(112 x 1.0)
(2 x 24)
(0.625 x 8.5)
@ 2.7 (9.0)
2896 x 31.75
63.5 x 711
L6x6x0.4375
15.875 x 216
(114 x 1.25)
(2.5 x 28)
(0.625 x 8.5)
@ 2.9 (9.5)
2896 x 31.75
57.2 x 711
L6x6x1
15.875 x 216
(114 x 1.25)
(2.25 x 28)
(0.625 x 8.5)
@ 2.9 (9.5)
2896 x 31.75
57.2 x 711
L6x6x0.375
15.875 x 216
(114 x 1.25)
(2.25 x 28)
(0.625 x 8.5)
@ 2.9 (9.5)
2794 x 22.225
44.5 x 508
L3.5x2.5x0.25
9.525 x 152
(110 x 0.875)
(1.75 x 20)
(0.375 x 6)
@ 4 (13)
2845 x 25.4
50.8 x 508
L3.5x2.5x0.5
15.875 x 216
24
Bearing
Stiffener
Dimensions,
mm (in)
25 x 254
(1.0 x 10)
25 x 254
(1.0 x 10)
25 x 279
(1.0 x 11)
25 x 305
(1.0 x 12)
34.9 x 305
(1.375 x 12)
25 x 254
(1.0 x 10)
25 x 254
23.13
5, 91 (300),
13.33
5, 91 (300),
16.67
5, 91 (300),
20.00
(112 x 1.0)
(2.0 x 20)
2896 x 31.75
(114 x 1.25)
57.2 x 660
(2.25 x 26)
L6x6x0.4375
2896 x 31.75
(114 x 1.25)
57.2 x 610
(2.25 x 24)
L6x6x1
2896 x 31.75
(114 x 1.25)
57.2 x 610
(2.25 x 24)
L6x6x0.375
(0.625 x 8.5)
@ 2.7 (9.0)
15.875 x 216
(0.625 x 8.5)
@ 2.9 (9.5)
15.875 x 216
(0.625 x 8.5)
@ 2.9 (9.5)
15.875 x 216
(0.625 x 8.5)
@ 2.9 (9.5)
(1.0 x 10)
25 x 279
(1.0 x 11)
25 x 305
(1.0 x 12)
34.9 x 305
(1.375 x 12)
Figure 31. Simplified Framing Plan, Balanced Two-Span, 4-Girder, 305-m (1000-ft) Radius Bridge.
Figure 32. Simplified Framing Plan, Balanced Two-Span, 4-Girder, 91-m (300-ft) Radius Bridge, R/L
= 13.33.
The unbalanced, two-span bridges had span lengths of 68.6 m (225.0 ft) and 34.3 m (112.5 ft). Prismatic
sections were again used for the initial erection studies.
25
Table 7 details the proportions of all steel superstructure components for the balanced, two-span, four-and
five-girder bridge designs. Figure 33 and Figure 34 show some representative four-girder framing plans.
26
Bridge (No. of
Girders;
Radius, m (ft);
R/L Ratio
4, 305 (1000),
35.59
4, 198 (650),
23.13
4, 91 (300),
13.33
4, 91 (300),
16.67
4, 91 (300),
20.00
5, 305 (1000),
35.59
5, 198 (650),
23.13
5, 91 (300),
13.33
5, 91 (300),
16.67
5, 91 (300),
20.00
Table 7. Unbalanced, Two-Span Bridge Proportions
Transverse
Stiffener
Dimensions,
Web
Flange
mm (in) and
Dimensions,
Dimensions,
Cross Frame
Spacings, m
mm (in)
mm (in)
Members
(ft)
2794 x 22.225
50.8 x 559
L3.5x2.5x0.25
9.525 x 152
(110 x 0.875)
(2 x 22)
(0.375 x 6)
@ 4 (13)
2845 x 25.4
50.8 x 660
L3.5x2.5x0.5
15.875 x 216
(112 x 1.0)
(2 x 26)
(0.625 x 8.5)
@ 2.7 (9.0)
2896 x 31.75
57.2 x 711
L6x6x0.4375
15.875 x 216
(114 x 1.25)
(2.25 x 28)
(0.625 x 8.5)
@ 2.9 (9.5)
2896 x 31.75
57.2 x 610
L6x6x1
15.875 x 216
(114 x 1.25)
(2.25 x 24)
(0.625 x 8.5)
@ 2.9 (9.5)
2896 x 31.75
57.2 x 610
L6x6x0.375
15.875 x 216
(114 x 1.25)
(2.25 x 24)
(0.625 x 8.5)
@ 2.9 (9.5)
2794 x 22.225
44.5 x 508
L3.5x2.5x0.25
9.525 x 152
(110 x 0.875)
(1.75 x 20)
(0.375 x 6)
@ 4 (13)
2845 x 25.4
50.8 x 508
L3.5x2.5x0.5
15.875 x 216
(112 x 1.0)
(2 x 20)
(0.625 x 8.5)
@ 2.7 (9.0)
2896 x 31.75
57.2 x 660
L6x6x0.4375
15.875 x 216
(114 x 1.25)
(2.25 x 26)
(0.625 x 8.5)
@ 2.9 (9.5)
2896 x 31.75
50.8 x 610
L6x6x1
15.875 x 216
(114 x 1.25)
(2 x 24)
(0.625 x 8.5)
@ 2.9 (9.5)
2896 x 31.75
50.8 x 610
L6x6x0.375
15.875 x 216
(114 x 1.25)
(2 x 24)
(0.625 x 8.5)
@ 2.9 (9.5)
Bearing
Stiffener
Dimensions,
mm (in)
25 x 254
(1.0 x 10)
25 x 254
(1.0 x 10)
25 x 279
(1.0 x 11)
25 x 305
(1.0 x 12)
25 x 254
(1.0 x 10)
25 x 254
(1.0 x 10)
25 x 254
(1.0 x 10)
25 x 279
(1.0 x 11)
25 x 254
(1.0 x 10)
25 x 229
(1 x 9)
Figure 33. Simplified Framing Plan, Unbalanced Two-Span, 4-Girder, 305-m (1000-ft) Radius
Bridge.
27
Figure 34. Simplified Framing Plan, Unbalanced Two-Span, 4-Girder, 91-m (300-ft) Radius Bridge,
R/L = 13.33.
Single-span bridges were statically analyzed following the procedure outlined in previous submittals for
PennDOT (Linzell et al. 2008) for the following four construction scenarios: (1) paired girder erection
placing the interior girders first; (2) paired girder erection placing the exterior girders first; (3) single girder
erection that placed the interior girder first; and (4) single girder erection that placed the exterior girder
first. The construction techniques were selected based on common erection procedures used by contractors
to erect horizontally curved, steel, I-girder bridges. Paired girder erection for the five-girder bridges
involved placing the two interior girders first, then placing the middle girder, and finally placing the two
exterior girders last. The reverse was done for paired girder erection by placing the exterior girders first. All
analyses involved applying dead load to the components of the structure that were erected during the
current phase of construction. The final stage for all construction scenarios involved placing the entire slab
on the structure. Figure 35 through Figure 40 detail the analyzed construction sequences for a
representative single-span, four-girder, 305-m (1000-ft) radius bridge undergoing paired and single girder
erection. Figure 41 through Figure 48 detail analyzed construction sequences for a representative singlespan, five-girder, 305-m (1000-ft) radius bridge undergoing paired and single girder erection.
Figure 35. Stage 1 of Construction for Paired Girder (Inner Girders Placed First) Erection of SingleSpan, 4-Girder, 305-m (1000-ft) Radius Bridge.
28
Figure 37. Stage 1 of Construction for Single Girder (Inner Girder Placed First) Erection of SingleSpan, 4-Girder, 305-m (1000-ft) Radius Bridge.
29
30
The two equal span bridges were analyzed using the same four construction scenarios: (1) paired girder
erection placing the interior girders first; (2) paired girder erection placing the exterior girders first; (3)
single girder erection that placed the interior girder first; and (4) single girder erection that placed the
31
exterior girder first. The first stages of construction involved placing the girders from Abutment 1 to 18.1
m (59.3 ft) beyond the pier. The next set of stages involved placing the girder from 18.1 m (59.3 ft) beyond
the pier to Abutment 2. The final stages of construction involved placing the slab on the structure in three
different stages. The slab was placed in the negative moment regions first, and then the final two stages
involved placing the concrete in the positive moment regions. Figure 49 through Figure 52 show the
studied construction sequences for a representative two-equal-span, four-girder, 305-m (1000-ft) radius
bridge. Figure 53 shows the modeled slab pour sequence for the two-equal-span, four- and five-girder,
305-m (1000-ft) radius bridges.
Figure 49. Stage 1 of Construction for Paired Girder (Inner Girders Placed First) Erection of Two
Equal Span, 4-Girder, 305-m (1000-ft) Radius Bridge.
32
Figure 53. Slab Pour Sequence for Two Equal Span, 4- and 5-Girder, 305-m (1000-ft) Radius
Bridges.
The two-unequal-span bridges were analyzed using the same four aforementioned construction scenarios.
The first stages of construction involved placing the girders from Abutment 1 to 8.6 m (28.1 ft) beyond the
pier. The next set of stages involved placing the girders from 8.6 m (28.1 ft) beyond the pier to Abutment
2. The final stages of construction involved placing the slab on the structure in three different stages.
Figure 54 through Figure 57 show the studied construction sequences for a two-unequal-span, four-girder,
305-m (1000-ft) radius bridge. Figure 58 shows the modeled slab pour sequence for the two-unequal-span,
four- and five-girder, 305-m (1000-ft) radius bridges.
Unequal Spans, 4-Girder, 305-m (1000-ft) Radius Bridge.
33
Figure 58. Slab Pour Sequence for Two Unequal Spans, 4- and 5-Girder, 305-m (1000-ft) Radius
Bridge.
The initial study involving structures with more than two spans involved examining the effects of erecting
the girders on the response of a single structure, Structure #7A, using two methods: from the inner to outer
radius of curvature, and from the outer to inner radius of curvature. Its geometry and proportions are
described in detail in previous submittals to PennDOT that summarized how the structure was monitored
during construction and how it was used to assist with developing the modeling procedure (Linzell et al.
2003; Hiltunen et al. 2004; Linzell et al. 2006; Linzell et al. 2008). The structure is six spans, composed of
five girders; it is divided into 2 three-span continuous units, and one of those units containing Spans 4
through 6 was examined. A simplified framing plan of the examined spans is shown in Figure 59.
34
Figure 59. Structure 7A Framing Plan.
35
5.2.1.2
Initial Studies - Results
Results from static, sequential analyses of the initial group of single- and two-span curved bridges were examined
statistically to identify preferred erection sequencing approaches from the previously discussed group of examined
erection sequencing options. Preferred sequences were identified based on deflections from the analyses, since
previous results for this study and other published work indicated that during construction, bridge performance is
largely controlled by stiffness and not strength (Linzell et al. 2003; Hiltunen et al. 2004; Linzell et al. 2006; Linzell
et al. 2008; Galambos et al. 1996). It should be noted that for the initial group of bridges that were studied, top and
bottom flange stresses were never observed to reach yield.
Deflection results and discussions are presented in perceived order of importance to contractors that erect the
structures. Vertical deflection is classified as the deflection direction with the highest importance, followed by radial
deflections, and finally tangential deflections, which varied slightly for different erection methods. Figure 60 details
the three deflection directions and boundary condition orientations. Deflections were obtained for each node along
the bottom flange of each girder for each stage of construction. Depending on the span type and the number of
girders, deflection results were generated for a minimum of 5 construction stages and a maximum of 13 stages. Due
to the large number of generated deflection values, only the paired construction deflection results for fascia Girders
1 and 4 for the single -span, four-girder, 305-m (1000 ft.) radius bridge are presented herein.
Figure 60. Plan View Detailing Deflection Directions.
Representative deflections for one of the thirty single- and two-span structures that were examined are presented.
After presenting these representative results, a discussion of the statistical methods used to determine the influence
of certain parameters on construction response is provided.
Figure 61 to Figure 63 show interior girder G1 vertical, radial, and tangential bottom flange deflections for the
paired inner and paired outer construction options, respectively, for the single-span, four-girder, 305-m (1000 ft.)
radius bridge. Stage 1 of the paired inner construction method corresponds to the erection of G1 and G2. Stage 2
36
corresponds to the addition of G3 and G4. For paired outer construction, Stage 1 is not shown because G1 had not
been erected. Stage 2 of the paired outer construction method corresponds to the addition of G1 and G2 to the
structure when G3 and G4 are in place. Stage 3 of both the paired inner and paired outer construction methods
corresponds to the placement of the entire slab on the structure.
Figure 61 shows that Stage 3 of the paired inner construction method resulted in the largest vertical deflection. The
figure also shows that during paired inner construction, the placement of G3 and G4 (Stage 2 paired inner) decreased
the vertical deflection for G1, which demonstrates the benefits of adding additional stiffness to the superstructure.
Figure 62 indicates that all radial deflections are nearly equal to zero at Abutment 1 (see Figure 60) due to the
restraint of translations normal to the plane of the superstructure and in the plane of the superstructure (―pinned‖
supports). Abutment 2 also restrains translations normal to the plan (―roller‖ supports) and, as a result, maximum
radial displacements occur at Abutment 2 for all stages of paired construction. The largest radial deflections for G1
occurred for paired inner construction Stages 2 and 3, which shows that the initial placement of G1 and G2 increases
the magnitude of the G1 radial deflection in subsequent stages. The radial deflection of G1 decreases significantly at
the cross frame locations for Stages 2 and 3 of the paired outer construction method.
Figure 63 shows that tangential deflections for G1 are consistently smaller in magnitude than the radial and vertical
deflections. The cumulative effect of the three stages of the paired inner construction method results in larger
tangential deflections for G1 when compared to the tangential deflections from the two stages of the paired outer
construction method.
Distance Along Girder from Abut. 1, m
0
10
20
30
40
50
60
1.0
7.6
0.0
-12.4
-1.0
-2.0
-52.4
-72.4
-3.0
Deflection, mm
Deflection, in.
-32.4
-92.4
-4.0
-112.4
-5.0
-132.4
-6.0
-152.4
0
50
100
150
200
Distance Along Girder from Abut. 1, ft.
Stage 1 Paired Inner
Stage 3 Paired Outer
Cross Frame Locations
Figure 61. Girder 1 Bottom Flange Vertical Deflections for Single-Span, 4-Girder, 305m (1000 ft.) Radius
Bridge.
37
10
20
30
40
50
60
8.0
198.4
6.0
148.4
4.0
98.4
2.0
48.4
0.0
-1.6
-2.0
-51.6
-4.0
Deflection, mm
Deflection, in.
0
-101.6
0
50
100
150
200
Figure 62. Girder 1 Bottom Flange Radial Deflections for Single-Span, 4-Girder, 305m (1000 ft.) Radius
Bridge.
38
10
20
30
40
50
60
1
24.9
0.8
19.9
0.6
14.9
0.4
9.9
0.2
4.9
0
Deflection, mm
Deflection, in.
0
-0.1
-0.2
-5.1
0
50
100
150
200
Bridge.
Figure 64 to Figure 66 show vertical, radial, and tangential bottom flange deflection values for paired inner and
paired outer construction approaches for exterior girder G4 of the single-span, four-girder, 305-m (1000 ft.) radius
bridge. In similar fashion to what was shown for certain stages for G1, Stage 1 for the paired inner construction
method is not presented because G4 had not yet been erected. To reiterate, Stage 2 of the paired inner construction
method corresponds to the addition of G3 and G4 to the structure when G1 and G2 are in place. Stage 1 of the paired
outer construction method corresponds to the erection of G3 and G4. Stage 2 of the paired outer construction method
corresponds to the addition of G1 and G2 to the structure. Stage 3 of both the paired inner and paired outer
construction methods corresponds to the placement of the entire slab on the structure.
Figure 64 shows that when maximum vertical deflections for G4 for both paired construction methods are compared,
Stage 3 of the paired outer construction method results in the largest vertical deflections. The cumulative effect of
placing G4 during Stage 1 of the paired outer construction method results in larger vertical deflections than waiting
to erect G4 during Stage 2 of the paired inner construction method.
Boundary conditions existed at the abutments for G4 that were similar to those discussed for G1. Therefore, Figure
65 shows that the maximum radial deflections for G4 occurred at Abutment 2 for all stages of paired construction.
The figure also shows that Stage 3 of the paired outer construction method results is the largest radial deflection for
G4.
Figure 66 shows smaller tangential deflections than vertical or radial deflections for G4, which is similar to the
results for G1. The figure also shows the maximum tangential deflections for Stage 3 of the paired outer
construction method.
39
10
20
30
40
50
60
2.0
43.6
0.0
-6.4
-2.0
-56.4
-4.0
-106.4
-6.0
-156.4
-8.0
-206.4
-10.0
-256.4
-12.0
-306.4
-14.0
-356.4
-16.0
Deflection, mm
Deflection, in.
0
-406.4
0
50
100
150
200
Figure 64. Girder 4 Bottom Flange Vertical Deflections for Single-Span, 4-Girder, 305m (1000 ft.) Radius
Bridge.
40
10
20
30
40
50
60
10.0
247.6
8.0
197.6
6.0
147.6
4.0
97.6
2.0
47.6
0.0
-2.4
-2.0
-52.4
-4.0
-102.4
-6.0
Deflection, mm
Deflection, in.
0
-152.4
0
50
100
150
200
Bridge.
41
0
10
20
30
40
50
60
2.5
60.0
50.0
40.0
1.5
30.0
1.0
Deflection, mm
Deflection, in.
2.0
20.0
0.5
10.0
0.0
0.0
0
50
100
150
200
Figure 66. Girder 4 Bottom Flange Tangential Deflections for Single-Span, 4-Girder, 305m (1000 ft.) Radius
Bridge.
An analysis of variance (ANOVA) procedure was selected to examine the influence of not only the construction
method, but also the number of girders, radius, unbraced length, and span type on maximum radial, tangential, and
vertical deflection values for the 120 one-and two-span bridge finite element models that were constructed (30
parametric bridges analyzed using 4 different construction methods). A general linear ANOVA model was selected in
Minitab (Minitab 2007) because it can test the influence of a number of different parameters on one response variable
(deflection). The three dependant variables were the vertical, radial, and tangential deflections. The aforementioned
five independent variables and their different levels as included in the ANOVA are shown in Table 8.
Table 8. Independent ANOVA Variables.
Variable
Construction Method
Number of Girders
Radius
Unbraced Length
Span Type
Levels
Single-Inner, Single-Outer, Paired-Inner, Paired Outer
4, 5
91m (300’), 198m (650’), 305m (1000’)
4.6m (15.0’), 5.3m (17.3’), 6.9m (22.5’), 9.6m (28.1’)
Single, Two Balanced, Two Unbalanced
To facilitate the ANOVA analysis, the radius and unbraced length independent variables were combined into a
single independent variable set, as shown in Table 9.
Table 9. R/L Ratios Used in ANOVA Analysis.
Radius, m (ft.)
Unbraced Length, m (ft.)
R/L Ratio
91 (300)
4.6 (15.0)
20.00
91 (300)
5.3 (17.3)
17.34
91 (300)
6.9 (22.5)
13.33
198 (650)
9.6 (28.1)
23.13
305 (1000)
9.6 (28.1)
35.59
42
Similar full statistical models were used for the ANOVA analyses to examine the influence of previously discussed
independent variable combinations on deflections. A full ANOVA statistical model includes the largest list of
variables that could be influencing the deflection response variable, and this full ANOVA statistical model was first
tested to determine which variables and which variable combinations had the largest influence on the resulting
deflections. As an example, Equation 1shows the results from the full ANOVA examination of independent
variable influence on vertical deflections generated during construction. The equation indicates that there are 4
main-effects terms, 6 two-way interaction terms, 3 three-way interaction terms, and 1 four-way interaction term for
the full, vertical deflection ANOVA model. Main-effect terms are multiplied together to create interaction terms,
which are included in the statistical model to determine if the combination of any of the main-effect terms is
influencing the deflection response. All four-way interaction terms are typically dropped from ANOVA analyses
and were done so here for all deflections.
Deflection = (Span + No. of Girders + Construction Method + R/L Ratio) + (Span * No. of Girders + Span *
Construction Method + Span * R/L Ratio + No. of Girders * Construction Method + No. of Girders * R/L Ratio +
Construction Method * R/L Ratio) + (Span * No. of Girders * Construction Method + Span * No. of Girders * R/L
Ratio + No. of Girders * Construction Method * R/L Ratio) + (Span * No. of Girders * Construction Method * R/L
Ratio)
Equation 1
In an attempt to further simplify the statistical models that were examined, a full ANOVA was performed using
Equation 1, with the last term removed, to ascertain which of the independent variables and which of their
combinations had the most significant effects on vertical deflections. Results from the ANOVA can be summarized
as shown in
43
Table 10. ANOVA Results for Initial Vertical Deflection Statistical Model.
Seq SS, m2
(ft.2)
0.12023
(1.29418)
0.02526
(0.27187)
0.43598
(4.69282)
0.39616
(4.26423)
0.00359
(0.03863)
0.03399
(0.36583)
0.08496
(0.91447)
Adj SS, m2
(ft.2)
0.12023
(1.29418)
0.02526
(0.27187)
0.43598
(4.69282)
0.39616
(4.26423)
0.00359
(0.03863)
0.03399
(0.36583)
0.08496
(0.91447)
Adj MS,
m2 (ft.2)
0.06012
(0.64709)
0.02526
(0.27187)
0.14533
(1.56427)
0.09904
(1.06606)
0.00180
(0.01932)
0.00566
(0.06097)
0.01062
(0.11431)
0.01648
(0.17740)
0.01648
(0.17740)
0.00549
(0.05913)
0.02774
(0.29861)
0.02680
(0.28849)
0.02774
(0.29861)
0.02680
(0.28849)
6
0.02926
(0.31492)
0.02926
(0.31492)
0.00694
(0.07465)
0.00223
(0.02404)
0.00488
(0.05249)
8
0.02368
(0.25380)
0.02368
(0.25380)
12
0.01404
(0.15110)
0.12814
(1.37928)
1.36620
(14.70563)
Source
DF
Span
2
No. of Girders
1
Construction Method
3
R/L Ratio
4
Span*No. of Girders
2
Span*Construction
Method
6
Span*R/L Ratio
8
No. of
Girders*Construction
Method
3
No. of Girders*R/L Ratio
4
Construction
Method*R/L Ratio
Span*No. of
Method
Span*No. of Girders*R/L
Ratio
No. of
Method*R/L Ratio
12
Error
48
Total
119
F
P
22.52
0.000
9.46
0.003
54.44
0.000
37.10
0.000
0.67
0.515
2.12
0.068
3.98
0.001
2.06
0.118
2.60
0.048
0.84
0.613
1.83
0.114
0.00295
(0.03172)
1.10
0.377
0.01404
(0.15110)
0.00117
(0.01250)
0.44
0.939
0.12814
(1.37928)
0.00267
(0.02874)
S = 0.05167 (0.16951)
R2 = 90.62%
R2 (adj) = 76.75%
.
The DF column in the table represents the degrees of freedom for each factor, which indicates the amount of
information the ANOVA statistical model factors provides for determining values for vertical deflection. The
sequential sum of squares (Seq SS) measures the amount of variation in the deflection response explained by adding
terms sequentially to the statistical model in the order listed in the table. The adjusted sum of squares (Adj SS)
measures the amount of additional variation in deflection explained by a specific factor, given that all other terms
are already in the statistical model. It can be noted that the sequential sum of squares values equaled the adjusted
sum of squares values for the vertical deflection, which shows that the order of the terms in the statistical model did
not affect the results. The adjusted mean of squares (Adj MS) is calculated by dividing the adjusted sum of squares
by the degrees of freedom. The S value, which comes from estimating the variance in the data and is calculated by
taking the square root of the adjusted mean squares of the error, represented the standard distance the data values fell
from any fitted values, with S being measured in deflection units, m (ft.). The better the model predicted vertical
deflection, the lower the S value. The R2 term, termed the coefficient of determination, represents the amount of
44
variation in the observed vertical deflections as a function of the independent variables (span, number of girders,
construction method, and R/L ratio). The adjusted R2 term was a modified R2 value adjusted for the number of terms
in the statistical model. The final columns in the table listed the F- and P-factors, respectively. Both represent
standard statistical tests, with the F-values found by dividing the adjusted mean squares by the adjusted mean
squares of the error term, and then being used as inputs to determine the P-values. The P-values determined whether
or not an independent variable or variable combination was a significant influence on determining vertical
deflections. P-values less than or equal to 0.05 correspond to variables or variable combinations that significantly
affected vertical deflections.
45
Table 10. ANOVA Results for Initial Vertical Deflection Statistical Model.
Seq SS, m2
Adj SS, m2
Adj MS,
2
Source
DF
(ft. )
(ft.2)
m2 (ft.2)
F
0.12023
0.12023
0.06012
Span
2
22.52
(1.29418)
(1.29418)
(0.64709)
0.02526
0.02526
0.02526
No. of Girders
1
9.46
(0.27187)
(0.27187)
(0.27187)
0.43598
0.43598
0.14533
Construction Method
3
54.44
(4.69282)
(4.69282)
(1.56427)
0.39616
0.39616
0.09904
R/L Ratio
4
37.10
(4.26423)
(4.26423)
(1.06606)
0.00359
0.00359
0.00180
Span*No. of Girders
2
0.67
(0.03863)
(0.03863)
(0.01932)
Span*Construction
0.03399
0.03399
0.00566
6
2.12
Method
(0.36583)
(0.36583)
(0.06097)
0.08496
0.08496
0.01062
Span*R/L Ratio
8
3.98
(0.91447)
(0.91447)
(0.11431)
No. of
0.01648
0.01648
0.00549
3
2.06
(0.17740)
(0.17740)
(0.05913)
Method
0.02774
0.02774
0.00694
4
2.60
(0.29861)
(0.29861)
(0.07465)
Construction
0.02680
0.02680
0.00223
12
0.84
Method*R/L Ratio
(0.28849)
(0.28849)
(0.02404)
Span*No. of
0.00488
0.02926
0.02926
6
(0.05249)
1.83
(0.31492)
(0.31492)
Method
Span*No. of Girders*R/L
0.02368
0.02368
0.00295
8
1.10
Ratio
(0.25380)
(0.25380)
(0.03172)
No. of
0.01404
0.01404
0.00117
12
0.44
(0.15110)
(0.15110)
(0.01250)
Method*R/L Ratio
0.12814
0.12814
0.00267
Error
48
(1.37928)
(1.37928)
(0.02874)
1.36620
Total
119
(14.70563)
S = 0.05167 (0.16951)
R2 = 90.62%
2
R (adj) = 76.75%
P
0.000
0.003
0.000
0.000
0.515
0.068
0.001
0.118
0.048
0.613
0.114
0.377
0.939
Examination of the results from the table indicate that the following independent variables or variable combinations
had significant effect on vertical deflections: (1) Span; (2) Number of Girders; (3) Construction Method; (4) R/L
Ratio; (5) Span * R/L Ratio; and (6) Number of Girders * R/L Ratio. As a result of these findings, a revised vertical
deflection ANOVA statistical model was developed that included these significant factors only. This model appears
as follows:
Vertical Deflection = Span + No. of Girders + Construction Method +
R/L Ratio + No. of Girders * R/L Ratio + Span * R/L Ratio
Equation 2
46
A second ANOVA was performed to ascertain if additional revisions to the vertical deflection statistical model were
necessary. Results from analyses of this model are shown in Table 11. They indicate that the P-value for the
combination of number of girders and R/L ratio was 0.052, and this term was removed from the final vertical
deflection statistical model, which is shown in Equation 3.
Table 11. ANOVA Results for Modified Vertical Deflection Statistical Model.
Seq SS, m2
Adj SS, m2
Adj MS,
2
Source
DF
(ft. )
(ft.2)
m2 (ft.2)
F
P
0.12023
0.12023
0.06012
Span
2
21.14
0.000
(1.29418)
(1.29418)
(0.64709)
0.02526
0.02526
0.02526
No. of Girders
1
8.88
0.004
(0.27187)
(0.27187)
(0.27187)
0.43598
0.43598
0.14533
Construction Method
3
51.10
0.000
(4.69282)
(4.69282)
(1.56427)
0.39616
0.39616
0.09904
R/L Ratio
4
34.82
0.000
(4.26423)
(4.26423)
(1.06606)
0.02774
0.02774
0.00694
4
2.44
0.052
(0.29861)
(0.29861)
(0.07465)
0.08496
0.08496
0.01062
Span*R/L Ratio
8
3.73
0.001
(0.91447)
(0.91447)
(0.11431)
0.27587
0.27587
0.00284
Error
97
(2.96945)
(2.96945)
(0.03061)
1.36620
Total
119
(14.70563)
S = 0.05333 (0.17497)
R2 = 79.18%
2
R (adj) = 75.23%
Vertical Deflection = Span + No. of Girders + Construction Method + R/L Ratio + Span * R/L Ratio
Equation 3
The final ANOVA statistical model included four significant single variables and one significant two-way
interaction term that contained the and R/L ratio. To ascertain if this final statistical model was acceptable for all
future studies, a final ANOVA was performed, with results shown in
47
Table 12. All P-values are less than 0.05, which shows that the factors in the final statistical model are significant.
The adjusted R2 value is 73.82%, which shows that 73.82% of the variation in the vertical deflections is explained
by the final statistical model.
48
Table 12. ANOVA Results for Final Vertical Deflection Statistical Model.
Seq SS, m2
Adj SS, m2
Adj MS,
2
Source
DF
(ft. )
(ft.2)
m2 (ft.2)
F
P
0.12023
0.12023
0.06012
Span
2
20.00
0.000
(1.29418)
(1.29418)
(0.64709)
0.02526
0.02526
0.02526
No. of Girders
1
8.40
0.005
(0.27187)
(0.27187)
(0.27187)
0.43598
0.43598
0.14533
Construction Method
3
48.34
0.000
(4.69282)
(4.69282)
(1.56427)
0.39616
0.39616
0.09904
R/L Ratio
4
32.95
0.000
(4.26423)
(4.26423)
(1.06606)
0.08957
0.08957
0.01062
Span * R/L Ratio
8
3.53
0.001
(0.91447)
(0.91447)
(0.11431)
0.30361
0.30361
0.00301
Error
101
(3.26805)
(3.26805)
(0.03236)
1.36620
Total
119
(14.70563)
S = 0.05483 (0.17988)
R2 = 77.78%
2
R (adj) = 73.82%
49
Using the final ANOVA from Equation 3, the influence of the four studied construction sequencing methods on
fitted mean vertical deflections for the 120 initially studied one- and two-span bridges was examined. Results are
shown in Figure 67, which indicates that construction methods did influence vertical deflection. However, an
additional statistical test was conducted to determine which differences in the fitted mean vertical deflections shown
in the figure were statistically significant. The Tukey method (Minitab 2007) was selected to perform this
investigation. The Tukey method produced adjusted P-values for each pair of fitted means, which represents the
probability that important differences exist between a pair of fitted means when, at first glance, that difference does
not appear statistically significant. For example, Figure 67 indicates that the single inner and paired inner
construction methods result in similar fitted mean vertical deflection values. The Tukey adjusted P-value for the
difference between the single inner and paired inner construction methods is 0.9937, which indicates that this
difference was not statistically significant, and thus supports what the figure shows. In other words, no significant
difference in the fitted mean vertical deflection exists when Girder 1 is first placed individually, assuming it is
adequately braced, verses placing both Girders 1 and 2 first. Erecting the interior girder (girder with the smallest
radius) first and then sequentially erecting the remaining girders in the structure minimizes the vertical deflection of
the structure irrespective of whether a single girder is placed first or a pair of girder is placed first. Placing the
interior girders first limits the final vertical deflections by reducing the structure’s tendency to rotate towards the
exterior girder. The figure clearly shows that the single outer construction method results in larger vertical
deflections when compared to the other three methods via examination of the fitted means. The Tukey adjusted Pvalues for the differences between the paired inner, paired outer, and single inner construction methods and the
single outer construction method are 0.0000, 0.0000, 0.0000, respectively, which further substantiate what the figure
shows with respect to first erecting the outer girder. Again, girders in the bridge tend to rotate towards the exterior
girder, and placing the exterior girder first magnifies this effect. When the fitted mean magnitudes were compared,
the smallest mean was for the single inner construction method, which would tend to indicate that this would be a
preferred erection approach for the group of bridges that was initially studied.
Figure 67. Construction Method Influence on Fitted Mean Vertical Deflections.
A similar ANOVA procedure was used to examine the influence of construction sequencing on radial deflections
obtained from the 120 one- and two-span bridges that were initially studied. After developing the full ANOVA
50
statistical model and reducing it, the final statistical model selected to examine sequencing effects on radial
deflections appears as shown below:
Radial Deflection = Span + No. of Girders + Construction Method + R/L Ratio + Span * No. of Girders + Span *
R/L Ratio + Construction Method * R/L Ratio
Equation 4
Table 13 shows the ANOVA results for the final radial deflection statistical model. The table indicates that all
factors included in the model were significant (P-values less than 0.05). The adjusted R2 value for the final radial
deflection statistical model is 69.36%.
Table 13. ANOVA Results for Final Radial Deflection Model.
Seq SS, m2
Adj SS, m2
Adj MS,
2
Source
DF
(ft. )
(ft.2)
m2 (ft.2)
F
0.00766
0.00766
0.00383
Span
2
1.49
(0.08240)
(0.08240)
(0.04120)
0.00176
0.00176
0.00176
No. of Girders
1
0.68
(0.01894)
(0.01894)
(0.01894)
0.20241
0.20241
0.06747
Construction Method
3
26.24
(2.17876)
(2.17876)
(0.72625)
0.30886
0.30886
0.07721
R/L Ratio
4
30.03
(3.32450)
(3.32450)
(0.83113)
0.02900
0.02900
0.01450
Span * No. of Girders
2
5.64
(0.31216)
(0.31216)
(0.15608)
0.05882
0.05882
0.00735
Span * R/L Ratio
8
2.86
(0.63311)
(0.63311)
(0.07914)
Construction Method *
0.16662
0.16662
0.01389
12
5.40
R/L Ratio
(1.79353)
(1.79353)
(0.14946)
0.22371
0.22371
0.00257
Error
87
(2.40804)
(2.40804)
(0.02768)
0.99884
Total
119
(10.75144)
S = 0.05071 (0.16637)
R2 = 77.60%
2
R (adj) = 69.36%
51
P
0.231
0.410
0.000
0.000
0.005
0.007
0.000
Figure 68 shows the effect of construction method on the radial deflection, again via examination and comparison of
fitted means. The figure clearly shows that the single outer construction method results in larger radial deflection
fitted mean values when compared to the other three methods. The Tukey adjusted P-values for the differences
between the paired inner, paired outer, and single inner construction methods and the single outer construction
method are 0.0000, 0.0000, 0.0000, respectively. Erecting the exterior girder (girder with the largest radius) first and
then sequentially erecting the remaining girders in the structure individually leads to excessive radial deflections
when compared to the other three construction methods.
Figure 68. Construction Method Influence on Fitted Mean Radial Deflections.
52
16.0
400.8
14.0
350.8
12.0
300.8
10.0
250.8
8.0
200.8
6.0
150.8
4.0
100.8
2.0
Fitted Means Radial Deflection, mm
Fitted Means Radial Deflection, in.
ince the final ANOVA statistical model for radial deflections included an interaction term that involved the
construction method, the effects of this interaction term on fitted mean radial deflections also needed to be
examined. Figure 69 details the effects of a combination of construction method and R/L ratio on radial deflection
fitted mean values. For R/L ratios of 23.13 and 35.59, the single outer construction method results in fitted mean
radial deflections that are significantly different when compared to the other three construction methods at those R/L
ratios. The results indicate that placing the exterior girder first results in larger radial deflections that cannot be
counteracted by the larger cross-frame spacings for the R/L values of 23.13 and 35.59.
50.8
13.33
17.33
20.00
23.13
35.59
R/L Ratio
Paired Inner Method
Paried Outer Method
Single Outer Method
Single Inner Method
Figure 69. Fitted Means Radial Deflection Comparison for the Interaction of Construction Method and R/L
Ratio.
The final statistical model for tangential deflections appears as shown in Equation 5. ANOVA results for the final
model are shown in
53
Table 14. The table indicates that all factors included in the model were significant (P-values less than 0.05). The
adjusted R2 value for the final statistical model is 84.51%.
Tangential Deflection = Span + Construction Method + R/L Ratio + Span * Construction Method + Span * R/L
Ratio
Equation 5
54
Table 14. ANOVA Results for Final Tangential Deflection Model.
Seq SS, m2
Adj SS, m2
Adj MS,
2
Source
DF
(ft. )
(ft.2)
m2 (ft.2)
F
0.01913
0.01913
0.00957
Span
2
216.37
(0.20596)
(0.20596)
(0.10298)
0.00184
0.00184
0.00061
Construction Method
3
13.84
(0.01976)
(0.01976)
(0.00659)
0.00641
0.00641
0.00160
R/L Ratio
4
36.26
(0.06904)
(0.06904)
(0.01726)
Span*Construction
0.00151
0.00151
0.00025
6
5.69
Method
(0.01625)
(0.01625)
(0.00271)
0.00083
0.00083
0.00010
Span*R/L
8
2.34
(0.00890)
(0.00890)
(0.00111)
0.00424
0.00424
0.00004
Error
96
(0.04569)
(0.04569)
(0.00048)
0.03397
Total
119
(0.36560)
S = 0.00665 (0.02182)
R2 = 87.50%
2
R (adj) = 84.51%
P
0.000
0.000
0.000
0.000
0.024
Figure 70 shows the effect of construction method on tangential deflection via an examination of fitted means. The
figure shows that the single outer construction method results in larger tangential deflection fitted means when
compared to the other three methods. The Tukey adjusted P-values for the differences between the paired inner,
paired outer, and single inner construction methods fitted means and the single outer construction method are
0.0000, 0.0014, 0.0000, respectively.
Figure 70. Construction Method Influence on Fitted Mean Tangential Deflections.
55
The final ANOVA statistical model for tangential deflections also included an interaction term that involved the
construction method, similar to the final radial deformation statistical model. So, again, the effects of this interaction
term on fitted mean radial deflections were examined. Figure 71 details the effects of a combination of construction
method and span type (i.e. single span, two span) on tangential deflection fitted means. The figure shows that for all
4 construction methods, the single-span bridges experience mean tangential deflection values larger than those
observed for the two-span bridges, and that the two-span bridges do not differ significantly with respect to
construction sequencing effects.
3.5
3.0
75.4
65.4
2.5
55.4
2.0
45.4
1.5
Fitted Means Tangential Deflection, mm
Fitted Means Tangential Deflection, in.
85.4
35.4
1.0
25.4
Paired Inner
Paired Outer
Single Outer
Single Inner
Construction Method
Single Span
Two Equal Spans
Two Unequal Spans
Figure 71. Fitted Mean Tangential Deflection Comparison for the Interaction of Construction Method and
Span Type.
In summary, based on the ANOVA vertical, radial and tangential deflection statistical models of the 120 one- and
two-span bridges that were initially studied, it can be concluded that the single outer construction method generally
resulted in the largest deflections, irrespective of bridge geometry. However, of the other methods, it was not clear
which would be preferred for a given bridge framing plan. Therefore, it was of interest to expand upon the ANOVA
statistical work to assist with future parametric studies. Results from each of the one- and two-span bridge analyses
were examined in greater detail to assist with providing a more comprehensive list of recommendations related to
preferred construction methods.
Each bridge was examined to determine if one of the investigated construction methods could be clearly identified
as that which minimized all three possible deformation components (i.e. radial, tangential, and vertical).
Conversely, the models were also examined to identify a construction method that maximized those deformations.
For the single-span bridges, the focus was at the end of all stages of construction, while for the two-span bridges, the
focus was the field splice locations when only the first portion of a girder section was in place. Comparisons
between deflections for different stages of construction greater than 6.35 mm (0.25 in.) were deemed significant.
Results from these examinations are summarized in Table 15 through
56
Initial Studies.
No. of
Girders
4
4
4
4
4
5
5
5
5
Radius, m (ft.)
91 (300)
91 (300)
91 (300)
198 (650)
305 (1000)
91 (300)
91 (300)
91 (300)
198 (650)
R/L Ratio
20.00
17.33
13.33
23.13
35.59
20.00
17.33
13.33
23.13
5
305 (1000)
35.59
Method
Single Outer
Single Inner
Single Outer
Single Inner
Single Outer
Single Outer
Single Outer
Single Outer
Single Outer
Paired Outer
Single Outer
and are differentiated based on span number.
Table 15. Recommended Construction Methods for Single-Span Bridges, Initial Studies.
No. of
Girders
Radius, m (ft.) R/L Ratio
Method
4
91 (300)
20.00
Single Inner
Paired Outer, Paired Inner,
4
91 (300)
17.33
Paired Outer
Paired Inner (Min. Radial)
4
91 (300)
13.33
Single Inner (Min. Vertical)
4
198 (650)
23.13
4
305 (1000)
35.59
Paired Inner
5
91 (300)
20.00
Single Inner
5
91 (300)
17.33
5
91 (300)
13.33
Paired Inner
5
198 (650)
23.13
5
305 (1000)
35.59
Table 16. Construction Methods Not Recommended for Single-Span Bridges, Initial Studies.
No. of
Girders
Method
4
91 (300)
20.00
Single Outer
4
91 (300)
17.33
Single Outer
4
91 (300)
13.33
Single Outer
4
198 (650)
23.13
Single Outer
4
305 (1000)
35.59
Single Outer
5
91 (300)
20.00
Single Outer
5
91 (300)
17.33
Single Outer
5
91 (300)
13.33
Single Inner
Single Outer
5
198 (650)
23.13
Paired Outer
5
305 (1000)
35.59
Single Outer
57
Table 17. Recommended Construction Methods for Balanced Two-Span Bridges, Initial Studies.
No. of
Girders
Method
4
91 (300)
20.00
Paired Inner
4
91 (300)
17.33
Paired Inner
4
91 (300)
13.33
Paired Inner
4
198 (650)
23.13
Paired Outer
4
305 (1000)
35.59
Paired Outer
5
91 (300)
20.00
5
91 (300)
17.33
Paired Outer
Paired Outer (Min. Radial)
5
91 (300)
13.33
5
198 (650)
23.13
Paired Inner
5
305 (1000)
35.59
Paired Outer
Table 18. Construction Methods Not Recommended for Balanced Two-Span Bridges, Initial Studies.
No. of
Girders
Method
4
91 (300)
20.00
Single Outer
4
91 (300)
17.33
Single Outer
Single Outer
4
91 (300)
13.33
Paired Outer
4
198 (650)
23.13
Single Outer
4
305 (1000)
35.59
Single Outer
Single Outer
5
91 (300)
20.00
Single Inner
5
91 (300)
17.33
Single Outer
5
91 (300)
13.33
Single Outer
Single Outer
5
198 (650)
23.13
Single Inner
5
305 (1000)
35.59
Single Outer
Table 19. Recommended Construction Methods for Unbalanced Two-Span Bridges, Initial Studies.
No. of
Girders
Method
Paired Outer (Min. Radial)
4
91 (300)
20.00
Paired Inner (Min. Vertical)
4
91 (300)
17.33
Single Outer
4
91 (300)
13.33
Paired Outer
4
198 (650)
23.13
Paired Inner
4
305 (1000)
35.59
Paired Outer
5
91 (300)
20.00
Paired Inner
5
91 (300)
17.33
5
91 (300)
13.33
5
198 (650)
23.13
Paired Outer
5
305 (1000)
35.59
Paired Inner
58
Table 20. Construction Methods Not Recommended for Unbalanced Two-Span Bridges, Initial Studies.
No. of
Girders
Method
4
91 (300)
20.00
Single Outer
4
91 (300)
17.33
Single Inner
4
91 (300)
13.33
Single Outer
4
198 (650)
23.13
Single Inner
4
305 (1000)
35.59
Single Outer
5
91 (300)
20.00
Single Outer
5
91 (300)
17.33
Single Outer
5
91 (300)
13.33
Single Outer
5
198 (650)
23.13
Single Outer
Paired Outer
5
305 (1000)
35.59
Single Outer
59
The results listed above indicate the following for the initial set of single- and two-span horizontally curved bridges
that were examined:
Construction methods that initiate with the inner (lowest radius) girder are preferred for single-span
horizontally curved structures, but a clear preference between erecting single girders and erecting girders in
pairs, from a strength and stiffness standpoint, is not apparent;
Paired construction methods, irrespective of the initial placement scheme relative to the center of curvature,
are preferred for two-span structures irrespective of span ratio; and
Regardless of span number and geometry, construction that initiates with a single outer girder is not
recommended from a strength and stiffness standpoint.
As was stated previously, the initial examination of effects of the aforementioned construction methods on the
response of structures containing more than two spans occurred via the study a single curved bridge, Structure #7A.
Three-dimensional models of the three-span continuous framing system comprised of Spans 4, 5, and 6 were
sequentially constructed in SAP 2000, and the aforementioned single-and paired-girder erection sequences were
completed computationally. Their effects on the structure were examined via the tangential, radial, and vertical
deflections that were induced during construction.
Single inner erection was examined by placing the smallest radius of curvature girder first from splice to splice,
followed by the next higher radius girder and so on until the entire superstructure was erected.
Single outer girder placement was investigated using a similar procedure, with the only difference being the
initiation of the erection from outer to inner girder. Outer and inner paired erection was also investigated in similar
fashion, with the paired girders and the fifth, single girder, being erected from splice to splice.
To evaluate and compare erection sequences for this structure, the final deformed shapes for the first and last girder
line erected for each were compared. Single girder approaches will be compared initially, followed by paired girder
approaches, and finally the four methods will be examined collectively.
Girder G5 was the first to be placed for the single outer erection procedure, while G1 was the first to be erected for
the single inner approach. The resulting final vertical and radial deformed shapes for the first and last girders erected
are shown in Figure 72 through Figure 75. Discontinuities evident in some of the plots were caused by SAPs
sequential analysis capabilities, which, at the time these analyses were completed, added new sections in their
undeformed positions to already deformed portions of the erected portions of the bridge. These discontinuities were
localized at the field splice locations and, since they affected all analyses that were completed, did not affect the
comparisons that were completed.
60
Arc Distance (m)
50
100
150
200
250
300
304.8
10
254
8
203.2
6
152.4
4
101.6
2
50.8
0
0
-2
Translation (mm)
Translation (in)
0
12
-50.8
-4
-101.6
0
200
400
600
800
1000
Arc Distance (ft)
G1 Outer-Inner
G5 Inner-Outer
Figure 72. Radial Deformations, Single Girder Erection. First Erected Girder.
Arc Distance (m)
0
50
100
150
200
250
152.4
2
101.6
0
-2
-50.8
-4
-101.6
-6
-152.4
-8
-203.2
-10
-254
-12
-304.8
0
200
400
600
800
1000
Arc Distance (ft)
G1 Outer-Inner
G5 Inner-Outer
Figure 73. Vertical Deformations, Single Girder Erection. First Erected Girder.
61
Translation (mm)
50.8
0
Translation (in)
300
4
Arc Distance (m)
50
100
150
200
250
300
38.1
1
25.4
0.5
12.7
0
0
-0.5
Translation (mm)
Translation (in)
0
1.5
-12.7
-1
-25.4
0
200
400
600
800
1000
Arc Distance (ft)
G5 Outer-Inner
G1 Inner-Outer
Figure 74. Radial Deformations, Single Girder Erection. Last Erected Girder.
Arc Distance (m)
50
100
150
200
250
300
50.8
1
25.4
0
0
-1
-25.4
-2
-50.8
-3
-76.2
-4
-101.6
-5
-127
-6
-152.4
-7
-177.8
0
200
400
600
800
1000
Arc Distance (ft)
G5 Outer-Inner
G1 Inner-Outer
Figure 75. Vertical Deformations, Single Girder Erection. Last Erected Girder.
62
Translation (mm)
Translation (in)
0
2
Examination of the results from these comparisons appeared to support conclusions drawn from the initial singleand two-span girder studies discussed earlier. These conclusions are that, if single-girder erection procedures are
used, choosing a single outer approach tends to result in greater displacements than a single inner approach.
Similar comparisons for the paired inner and outer approaches are shown in Figure 76 through Figure 79.
Arc Distance (m)
0
30
61
91
122
152
183
213
244
274
305
25.4
0.8
20.32
0.6
15.24
0.4
10.16
0.2
5.08
0
0
-0.2
-5.08
-0.4
-10.16
-0.6
-15.24
-0.8
-20.32
0
100
200
300
400
500
600
700
800
900
1000
Arc Distance (ft)
G1 Outer-Inner
G5 Inner-Outer
Figure 76. Radial Deformations, Paired Girder Erection. First Erected Girder.
63
Translation (mm)
Translation (in)
1
Arc Distance (m)
30
61
91
122
152
183
213
244
274
305
2
50.8
1
25.4
0
0
-1
-25.4
-2
-50.8
-3
-76.2
-4
-101.6
-5
-127
-6
-152.4
-7
Translation (mm)
Translation (in)
0
-177.8
0
100
200
300
400
500
600
700
800
900
1000
Arc Distance (ft)
G1 Outer-Inner
G5 Inner-Outer
Figure 77. Vertical Deformations, Paired Girder Erection. First Erected Girder.
Arc Distance (m)
30
61
91
122
152
183
213
244
274
305
25.4
0.4
20.32
0.3
15.24
0.2
10.16
0.1
5.08
0
0
-0.1
-5.08
-0.2
-10.16
-0.3
-15.24
-0.4
-20.32
0
100
200
300
400
500
600
700
800
900
1000
Arc Distance (ft)
G5 Outer-Inner
G1 Inner-Outer
Figure 78. Radial Deformations, Paired Girder Erection. Last Erected Girder.
64
Translation (mm)
Translation (in)
0
0.5
Arc Distance (m)
0
30
61
91
122
152
183
213
244
274
25.4
0.5
12.7
0
-0.5
-12.7
-1
-25.4
-1.5
-38.1
-2
-50.8
-2.5
-63.5
-3
-76.2
-3.5
-88.9
-4
Translation (mm)
0
Translation (in)
305
1
-101.6
0
100
200
300
400
500
600
700
800
900
1000
Arc Distance (ft)
G5 Outer-Inner
G1 Inner-Outer
Figure 79. Vertical Deformations, Paired Girder Erection. Last Erected Girder.
The paired girder comparisons indicate that the paired inner approach produced, in general, smaller deformations
than the paired outer approach. However, the differences in deformation magnitude were not nearly as dramatic as
for single girder comparisons, which again appear to support conclusions drawn from the single- and two-span
initial studies.
When all four studied erection procedures are compared, paired-girder erection procedures were preferred over
single-girder for the single three-span structure that was initially investigated. Again, these final conclusions appear
to support the conclusions obtained for the initial single- and two-span structure studies summarized in Table 15
through
65
Initial Studies.
No. of
Girders
4
4
4
4
4
5
5
5
5
Radius, m (ft.)
91 (300)
91 (300)
91 (300)
198 (650)
305 (1000)
91 (300)
91 (300)
91 (300)
198 (650)
R/L Ratio
20.00
17.33
13.33
23.13
35.59
20.00
17.33
13.33
23.13
5
305 (1000)
35.59
Method
Single Outer
Single Inner
Single Outer
Single Inner
Single Outer
Single Outer
Single Outer
Single Outer
Single Outer
Paired Outer
Single Outer
.
5.2.1.3
Representative Structure Studies - Background
Five bridges from the set described in the Final Design section were selected for additional verification with respect
to erection sequencing effect results. It was of interest to ensure that the recommended preliminary study erection
schemes were mimicked in the final design structure. The selected bridges were two-span structures having varying
radii and three-span bridges having balanced and unbalanced spans. These bridges were selected so they included
important parameters examined in the preliminary studies, such as extreme R/L ratio and incorporation of the
unbalanced spans. The radii for the three two-span, four-girder bridges selected were 91.4 m (300 ft), 198.1 m (650
ft), and 304.8 m (1000 ft), and the radius of the single balanced and the single unbalanced three-span, four-girder
bridges selected was 91.4 m (300ft). The unbalanced three-span bridge span ratio was 1:1.4. The designed girder
spacing and cross-frame spacing for the selected representative bridges were 3 m (10 ft) and 6.9 m (22.5 ft),
respectively. Consequently, three different R/L ratios (13.3, 28.9, and 44.4) were investigated. In addition, the deck
was designed following the deck dimensions used in the previous initial studies. The span length for the two-span
bridges and the balanced, three-span bridge was 68.6 m (225 ft), and the span lengths for the unbalanced, three-span
bridge were 48.1 m (157.5 ft) and 68.6 m (225 ft).
66
Table 21 lists the parameters of the selected representative bridges. Simplified framing plans for the selected
bridges are shown in Figure 80 to Figure 84.
67
Table 21: Selected Curved Bridge Erection Study Bridge Information
Bridge
No.
Radius of
Curvature, m
(ft)
CrossFrame
Spacing, m
(ft)
GirderSpacing,
m (ft)
Number
of
Spans
C3
91.4 (300)
6.9 (22.5)
3 (10)
2
C6
198.1 (650)
6.9 (22.5)
3 (10)
2
C9
304.8 (1000)
6.9 (22.5)
10
2
C10
91.4 (300)
6.9 (22.5)
3 (10)
3
C11
91.4 (300)
6.9 (22.5)
3 (10)
3
Span-Length, m
(ft)
68.6-68.6
(225-225)
68.6-68.6
(225-225)
68.6-68.6
(225-225)
68.6-68.6-68.6
(225-225-225)
48.1-68.6-68.6
(157.5-225-225)
Number of
Girder, m
(ft)
4
4
4
4
4
Figure 80: Simplified Framing Plan, Bridge C3: Two-Span, 4-Girder, 91.4 m Radius, R/L=13.3
68
Figure 83: Simplified Framing Plan, Bridge C10: Balanced, Three-Span, 4-Girder, 91.4 Radius, R/L=13.3
Figure 84: Simplified Framing Plan, Bridge C11: Unbalanced, Three-Span, 4-Girder, 91.4 Radius, R/L=13.3
69
Four erection scenarios were applied to the selected structures to reaffirm results from the preliminary studies by
investigating the effects of erection sequencing decisions on construction behavior. The four erection scenarios
were: 1) paired-girder erection placing the interior girders first; 2) paired-girder erection placing the exterior girders
first; 3) single-girder erection placing the interior girder first; and 4) single-girder erection placing the exterior girder
first. The behavior of the structures under self weight was analyzed using ABAQUS/Standard, with sequential
analysis being performed to examine the behavior for construction stages that involved girder erection and concrete
deck placement, with the final stage for the sequential analysis being deck placement. Figure 85 through Figure 88
detail the superstructure erection stages for the four scenarios for a representative two-span bridge. A similar
erection approach was used for the three-span bridges, with the number of erection stages differing due to an
increased number of splices. In addition to erection sequencing study, the effects of ―drop-in‖ erection on the
constructability of curved bridges were examined for balanced and unbalanced three-span structures having a severe
curvature. The ―drop-in‖ erection sequences for Bridges C-10 and C-11, the two bridges studied, are shown in
Figure 89 and Figure 90.
Stage-1
Stage-2
Stage-3
Stage-4
Stage-5
Stage-6
Figure 85: Stage 1 to Stage 6 for Paired-Girder (inner girders placed first) Erection of Two-Equal-Span, 4Girder bridges.
70
Stage-1
Stage-2
Stage-3
Stage-4
Stage-5
Stage-6
Figure 86: Stage 1 to Stage 6 for Paired-Girder (outer girders placed first) Erection of Two-Equal-Span, 4Girder Bridges.
71
Stage-1
Stage-2
Stage-3
Stage-4
Stage-5
Stage-6
Stage-7
Stage-8
Stage-9
Stage-10
Stage-11
Stage-12
Figure 87: Stage 1 to Stage 12 of Construction for Single-Girder (inner girder placed first) Erection of TwoEqual-Span, 4-Girder Bridges.
72
Stage-1
Stage-2
Stage-3
Stage-4
Stage-5
Stage-6
Stage-7
Stage-8
Stage-9
Stage-10
Stage-11
Stage-12
Figure 88: Stage 1 to Stage 12 of Construction for Single-Girder (outer girder placed first) Erection of TwoEqual-Span, 4-Girder Bridges.
73
Stage-1
Stage-2
Stage-3
Stage-4
Stage-5
Stage-6
Stage-7
Stage-8
Stage-9
Stage-10
Stage-11
Stage-12
74
Stage-13
Stage-14
Figure 89: Stage 1 to Stage 14 of Construction for “Drop-In” Erection of Bridge C-10.
Stage-1
Stage-2
Stage-3
Stage-4
Stage-5
Stage-6
Stage-7
Stage-8
Figure 90: Stage 1 to Stage 8 of Construction for “DropIn” Erection of Bridge C-11.
5.2.1.4
Representative Structure Studies - Results
Based on the results of the initial studies, vertical and radial deflections were classified as the two most important
indexes for evaluating the adequacy of various erection approaches. Therefore, vertical and radial deflections from
sequential analyses were compared herein to assist with the current evaluation. To compare the different erection
sequences, vertical and radial deflections where maximum values were anticipated to occur were compared. These
75
locations were at splice locations and at 0.4L, where L is the span of the girder, measured from the abutment for the
exterior spans and at mid-span for interior spans.
For these analyses, comparisons were made by examining non-dimensionalized deflections that consisted of
maximum deflections normalized with respect to maximums from the recommended preliminary study erection
scenario (paired inner). The maximum deflection was defined as the maximum deflection at the selected locations in
the girders for a given stage. Results are presented in Figure 91 through Figure 100, with the construction stages
corresponding to those for the paired-girder approach. As observed in the figures, vertical deflections were
decreased when paired-girder erection approaches were used, while radial deflections were not significantly
influenced between the different erection methods. These results confirmed the findings from the initial studies of
the final bridge designs that were examined. If enough crane capacity is available, paired-girder erection approaches
would be the preferred erection methods for the curved bridges examined in this study. More specifically, the paired
inner erection approach was preferred for curved bridges having severe curvature. Maximum vertical and radial
girder deflections at splice locations obtained from the drop-in effect study were all close to those from previous
results, which permitted the drop-in erection approach to be adopted as an alternative method. Consequently, those
deflections were found to be functions of R/L values, boundary conditions, and span lengths, but not the drop-in
effect. Therefore, if adequate control of the already erected sections can be maintained, the use of drop-in erection
sequencing should not positively or adversely affect bridge constructability for the bridges studied.
Figure 91: Ratio of Maximum Vertical Deflections for Bridge C3.
76
77
78
Figure 96: Ratio of Maximum Radial Deflections for Bridge C3.
79
80
5.2.2
5.2.2.1
Parametric Studies - Skewed
Representative Structure Studies – Background
Similar to the study of the representative curved bridges, a total of five representative skewed structures were
examined. One single-span bridge with severely skewed supports, two two-span bridges with varying skew angles,
and two triple-span bridges with balanced/unbalanced spans were selected to examine erection sequencing effects.
These bridges were selected in order to address all of the important parameters assumed to have an impact on
skewed bridge constructability. Single- and two-span bridges were chosen to investigate the effects of span
continuity. Varying the skew angle in two-span bridges was considered to examine the behavior of varying skews
during construction. Three-span bridges having balanced and unbalanced spans were selected to examine different
span-ratio effects. The supports in the single- and three-span bridges had a severe skew of 50 , and the two-span
bridges were skewed of 70 and 50 relative to their supports. The unbalanced three-span bridge had a span ratio of
1:1.4. The designed girder spacing and cross-frame spacing for the selected representative bridges was 3 m (10 ft)
and 7.8 m (25.7 ft), respectively. The span length for the single-span, two-span bridges and the balanced, three-span
bridge was 54.86 m (180 ft), and the span lengths for the unbalanced, three-span bridge were 39.24 m (128.75 ft)
and 54.86 m (180 ft). Table 22 lists the parameters of the selected representative bridges. Simplified framing plans
for the selected representative bridges are shown in Figure 101 through Figure 105.
Table 22: Information From Representative Bridges.
Bridge
No.
Skew Angle
S2
50
CrossFrame
Spacing, m
(ft)
7.8 (25.7)
S6
50
S8
GirderSpacing,
m (ft)
Number
of
Spans
3 (10)
1
7.8 (25.7)
3 (10)
2
70
7.8 (25.7)
3 (10)
2
S9
50
7.8 (25.7)
3 (10)
3
S10
50
7.8 (25.7)
3 (10)
3
81
Span-Length, m
(ft)
54.86 (180)
54.86-54.86
(180-180)
54.86-54.86
(180-180)
54.86-54.86-54.86
(180-180-180)
54.86-54.86-39.24
(180-180-128.75)
Number
of Girder,
m (ft)
4
4
4
4
4
Figure 101: Simplified Framing Plan of Bridge S2: Single-Span, 50 Skew.
Figure 102: Simplified Framing Plan of Bridge S6: Two-Span, 50 Skew.
Figure 103: Simplified Framing Plan of Bridge S8: Two-Span , 70 Skew.
Figure 104: Simplified Framing Plan of Bridge S9: Balanced, Three-Span, 50 Skew.
Figure 105: Simplified Framing Plan of Bridge S10: Unbalanced, Three-Span, 50 Skew.
82
Unlike curved bridges, there is no differentiation between ―inner or outer‖ placement of the girders for the skewed
bridges that are studied due to the lack of a center of curvature and the resulting identical span lengths of adjacent
bridge girders. Therefore, two erection scenarios were applied to the representative structures to investigate the
effects of erection sequencing on their behavior during construction: 1) paired girder erection; and 2) single girder
erection. Similar to the curved bridges that were examined, sequential analysis was performed using ABAQUS to
examine skewed structure behavior after each construction stage that included girder erection and concrete deck
placement. Figure 106 and Figure 107 show the construction stages for the two types of erection sequences that were
examined for representative two-span bridge. A similar approach for erection sequencing was used for the singlespan and balanced and unbalanced three-span bridges.
Stage-1
Stage-2
Stage-3
Stage-4
Stage-5
Stage-6
Figure 106: Stage 1 to Stage 6 of Construction for Paired-Girder Erection of Two Equal-Span, 4-Girder
Bridges.
83
Stage-1
Stage-2
Stage-3
Stage-4
Stage-5
Stage-6
Stage-7
Stage-8
Stage-9
Stage-10
Stage-11
Stage-12
Figure 107: Stage 1 to Stage 12 of Construction for Single-Girder Erection of Two-Equal-Spans, 4-Girder
Bridges.
84
The effects of ―drop-in‖ erection on the constructability of skewed bridges were also examined for balanced and
unbalanced three-span structures having a severe skew. The ―drop-in‖ erection sequences for Bridge S-9 and S-10
are shown in Figure 108.
Stage-1
Stage-2
Stage-3
Stage-4
Stage-5
Stage-6
Stage-7
Stage-8
Stage-9
Stage-10
Figure 108: Stage 1 to Stage 10 of Construction for “Drop-In” Erection of Bridge S-9 and Bridge S-10.
85
5.2.2.2
Representative Structure Studies - Results
As opposed to the curved bridges, lateral deformation of skewed girders during their erection is not significant.
Girder twist becomes important during the concrete deck pour for skewed bridges, after the whole steel
superstructure has been erected. Therefore, only vertical deflections from the sequential analyses were compared
herein to evaluate the influences of different erection scenarios. Vertical deflections were compared at 0.4L from the
abutment for the exterior spans and at mid-span for the interior spans, since these locations are where maximum
deflections were assumed to occur. In addition, similar the curved bridge study, for each stage of construction the
maximum deflections, normalized with respect to the maximum deflection for the paired erection scenario, were
examined. Results are shown in Figure 109 through Figure 113, with stage numbers representing the paired erection
cases. As can be inferred from the figures, vertical deflections did not change much between the two erection
techniques. For the skewed bridges as well as the curved bridges, maximum vertical and radial girder deflections at
mid-span and splice locations obtained from the drop-in effect study were all close to those from previous results,
which permitted the drop-in erection approach to be adopted as an alternative method. As a result, it is apparent that
drop-in erection can not significantly affect the constructability of the skewed structures.
Figure 109: Ratio of Maximum Vertical Deflections for Bridge S2, Single-Span, 50 Skew.
Figure 110: Ratio of Maximum Vertical Deflections for Bridge S6, Two-Span, 50 Skew.
86
Figure 111: Ratio of Maximum Vertical Deflections for Bridge S8, Two-Span, 70 Skew.
Figure 112: Ratio of Maximum Vertical Deflections For Bridge S9, Balanced Three-Span, 50 Skew.
Figure 113: Ratio of Maximum Vertical Deflections for Bridge S10, Unbalanced Three-Span, 50 Skew
87
5.2.3
Final Results and Discussion
The results from the preliminary and final curved bridge erection sequence studies completed herein indicated that,
should adequate crane capacity be available and stability concerns be mitigated, vertical deflections were decreased
when paired-girder erection approaches were employed. Radial deflections did not significantly change between the
different erection methods that were examined. Of the paired-girder approaches that were studied, paired inner
erection was preferred for curved bridges with severe curvature. In addition, it was observed from the results that
bridge deflections were mainly controlled by R/L values, boundary conditions, and span lengths, but not by the
―drop-in‖ effect. The observed rotations for sections prior to drop-in erection were small enough to allow for this
erection approach should adequate control of the previously erected sections be provided.
For the skewed structures that were studied, the results indicated very small differences between examined erection
techniques for all bridges that were examined. Although paired erection can result in slightly lower vertical
deformations compared to single-girder erection, in general there was no marked difference between the techniques.
Drop-in erection methods were also permissible for skewed bridges because of the small rotations of the previously
erected girders.
In summary, findings from the erection sequencing parametric studies for the curved and skewed bridges that were
examined included:
Curved
o Girder vertical deflections were decreased when paired girder erection methods were
used.
o Paired inner erection was preferred for structures with severe curvature.
o Drop-in erection would be an acceptable approach.
Skewed
o Erection methods examined herein did not show appreciable influence on skewed bridge
behavior.
5.3
Web-Plumbness
This section examines the effects of web-plumbness on bridge behavior during construction. According to section
5.23 of the Structural Welding Code (AWS, 2004), maximum out-of-straightness of 1 in 100 is allowed for the web
of the plate girders. Hence, a 1% web out-of-plumbness was introduced to representative bridges to investigate its
effects on the constructability of curved and skewed bridges. This out of plumbness was applied to the girders in a
fashion that increased any twist in the girders caused by curvature or global twisting of the structure during deck
placement. In the models, the out-of-plumbness was introduced by intentionally tilting whole girder webs by 1% of
the web depth (as shown in Figure 114), and the geometries of the cross frames were modified accordingly. The
same magnitude and direction for out-of-plumbness was maintained along the entire length of the girders.
88
a)
Curved Bridge Representative Plan
b) Skewed Bridge Representative Plan
c)
Section A-A Indicating Web-Out of Plumbness Orientation for Girders in Curved and Skewed
Bridges
Figure 114: Girder Web Out-Of-Plumbness Information.
5.3.1
5.3.1.1
Parametric Studies
Curved
In curved bridges, web out-of-plumbness can aggravate the effects of torsional moment due to the curvature. These
effects are influenced by the radius and cross-frame spacing. Consequently, the R/L value can be used as a key
parameter for this parametric study to ascertain web out-of-plumbness effects. Examination of the effects of webplumbness on curved bridges consisted of five representative bridges having varying radii and cross-frame spacings.
The radii of the selected curved bridges were 91.4 m (300 ft), 198.1 m (650 ft), and 304.8 m (1000 ft), and the crossframe spacings were 4.57 m (15 ft) and 6.86 m (22.5 ft). These parameters were selected so they included
moderately to severely curved bridges with different cross-frame spacings, resulting in a broad range of R/L values
(from 13.3 to 66.7). To investigate the effects of web-plumbness, a 1% web out-of-plumbness (as shown in Figure
114) was introduced to girder webs in a fashion that aggravated the curvature effects Sequential analysis was
89
performed to examine the effects at each stage, with the paired inner erection method being adopted for this study,
since it was considered the preferred erection approach. Table 23 lists parameters for the bridges that were examined.
Bridge
No.
C1
C3
C6
C7
C9
5.3.1.2
Table 23 Selected Curved Bridge Web-Plumbness Study Information.
CrossNumber
Radius of
GirderFrame
of Spans
Span-Length, ft
Curvature, ft
Spacing, ft
Spacing, ft
2
300
15
10
225-225
10
2
300
22.5
225-225
10
2
650
22.5
225-225
10
2
1000
15
225-225
10
2
1000
22.5
225-225
Number of
Girder, ft
4
4
4
4
4
Skewed
Five representative skewed bridges were also considered in order to investigate the effects of web-plumbness on
their construction behavior. Representative bridges included 1 single-span bridge with a 50 skew and four two-span
bridges with 50 and 70 skews and varying cross-frame spacing. The cross-frame spacings in the one-span bridge
were 7.8m (25.7 ft), and they were 4.57 m (15 ft) and 7.8 m (25.7 ft) in the two-span bridges. Similar to the study of
the curved structures, these parameters were selected so they included both moderately to severely skewed bridges
with different cross-frame spacings, and one- and two-span bridges were selected to examine the influence of
continuity. Sequential analysis was performed to simulate bridge behavior for each construction stage, and the
paired-girder erection method was again adopted. Table 24 lists parameters of the selected representative skewed
bridges.
Bridge
No.
S2
S5
S6
S7
S8
5.3.2
5.3.2.1
Table 24 Selected Skewed Bridge Web-Plumbness Study Information.
CrossNumber
Skew
Number
Frame
Girderof
Angle,
Span-Length, ft
of Girder,
Spacing,
Spacing, ft
Spans
degree
ft
ft
10
50
25.7
1
180
4
10
50
15
2
180-180
4
10
50
25.7
2
180-180
4
10
70
15
2
180-180
4
10
70
25.7
2
180-180
4
Results and Discussion
Curved
Similar to the erection sequencing study, vertical and radial deflections were primarily compared to examine the
effects of web-plumbness on construction response. At each stage, maximum girder vertical deflections and
maximum radial deflections at splice locations were examined for the out-of-plumb cases, with those maximum
deflections being normalized by the maximum deflections for the baseline case that had plumb webs. Comparisons
are shown in Figure 115 to Figure 124.
The results showed that the effects of web out-of-plumbness on both vertical and radial deflections were
insignificant. Not surprisingly, the change in vertical deflections due to web out-of-plumbness was negligible
(typically less than 1%). Although the change in ratios of radial deflections was larger than that for vertical
deflections, those deflection changes were small compared to the radial deflections due to curvature effects.
Furthermore, cumulative stresses in the cross frames were checked for two of the severely curved bridges (R=300’).
To examine out-of-plumbness effects on stresses in the structures, comparisons of maximum Von Mises stresses,
90
which account for the effects of axial, bending and shear stresses in the cross frames, are presented in Figure 125
and Figure 126. Similar findings were observed from the stress results. The changes in stresses in cross frames were
less than 1%. Hence, the effects of a 1% web out-of-plumbness on bridge constructability are considered small for
the structures that were studied.
91
92
93
94
95
Figure 125: Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge C1.
An additional investigation into the effects of unintentional web out-of-plumbness on the behavior of curved bridges
occurred via the examination of six times the allowable value (6%) for Bridge C3, a structure with critical R/L
values from those that were studied and one that showed slightly larger changes in deformations and stresses for 1%
out-of-plumbness when compared to the other structures that were examined (see Table 4). The 6% value was
applied to the girders in the same fashion as that used for 1% and vertical and radial deflections and Von Mises
stresses in the cross frames were compared to the baseline, web plumb case, to examine their effects. Comparisons
are shown in Figure 127 through Figure 129. As shown in these figures, with an extreme web out-of-plumbness of
6%, changes in deformations and stresses were 20% or less. This indicates that these levels of out-of-plumbness,
which are certainly a concern from a constructability perspective, do not impose a drastic change to curved bridge
behavior during construction and that global curvature effects dominate the construction behavior. It should also be
noted that web out-of-plumbness was shown to have a slightly larger influence on radial deformations than on
vertical deformations and cross frame stresses, irrespective of the out-of-plumbness magnitude.
96
Figure 127. Ratio of Maximum Vertical Deflections for Bridge C3, 6% Out-of-Plumbness.
Figure 128. Ratio of Maximum Radial Deflections for Bridge C3, 6% Out-of-Plumbness.
97
Figure 129 Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge C3, 6% Out-of-Plumbness.
5.3.2.2
Skewed
Vertical and radial deflections were also assumed to be important indices for examining the effects of webplumbness on skewed bridges. At each stage, the maximum vertical deflections in girders and the maximum radial
deflections at splice locations for web-plumb and web out-of-plumb structures were compared. All the deformation
values were normalized with respect to these results from the web-plumb bridges. Comparisons of deflections
between web-plumb and web out-of-plumb structures are shown in Figure 130 through Figure 139.
Figure 130: Ratio of Maximum Vertical Deflections for Bridge S2.
98
99
100
Figure 135: Ratio of Maximum Radial Deflections for Bridge S2.
101
Figure 138: Ratio of Maximum Radial Reflections for Bridge S7.
102
Similar to the study of curved bridges, the results showed that web out-of-plumbness had a very small impact on
vertical deformations, with effects being smaller than those on the curved bridges. For lateral deformations, as
shown in the figures, the web out-of plumbness caused a large increase in nondimensional ratios. As discussed in the
previous section, the large difference is caused by a slight increase in lateral deformation, and it not a significant
issue during the construction of these bridges. Cross-frame cumulative stresses were also checked for two of the
severely skewed bridges (50 skew). Comparisons of maximum Von Mises stresses in the cross frames are
presented in Figure 140 and Figure 141. Similar findings to the curved structures were observed, with changes in
cross-frames stresses being less than about 1%. Based on previously discussed results from this parametric study, it
was surmized that the effects of 1% web out-of-plumbness on skewed bridges during their construction were small.
Figure 140: Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge S5.
103
In similar fashion to the curved bridges, a 6% web out-of-plumbness was considered for Bridge S6, a structure
having the largest skew angle and cross frame spacing and one that was shown to have a slightly larger influence on
its behavior for 1% out-of-plumbness among those that were studied (Table 4). The 6% out-of-plumbness was
applied in the same fashion that was used for 1% and vertical and lateral girder deflections and Von Mises stresses
in the cross frames were compared to the baseline, web-plumb, case. Comparisons are shown in Figure 142 through
Figure 144.
Figure 142: Ratio of Maximum Vertical Deflections for Bridge S6, 6% Out-of-Plumbness..
104
Figure 143: Ratio of Maximum Lateral Deformations for Bridge S6, 6% Out-of-Plumbness.
Figure 144: Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge S6, 6% Out-of-Plumbness.
Based on the results, the 6% web out-of-plumbness in the skewed bridge could change vertical deflections of the
girders and stresses in the cross frames by a maximum of approximately25% when compared to the web-plumb
case, values that are slightly larger than those seen for the curved bridge with similar web out-of-plumbness. For
girder lateral deformations, however, a sixfold increase in web out-of-plumbness resulted in an increase in their
values by approximately six times. This demonstrates that, unlike for curved bridges where global curvature
dominates the behavior, for skewed structures the effects of web out-of-plumbness are more pronounced and, if
accounted for during erection, can be taken advantage of to ensure a web-plumb position at the completion of
construction.
5.3.2.3
Summary
This section examined the effects of web-plumbness on bridge constructability for curved and skewed structures by
comparing girder deflections and cross-frame stresses. A 1% web out-of-plumbness was applied to all bridge girders
in a fashion that increased any twist in the girders caused by curvature or global twisting of the structure during deck
placement. Results showed that no appreciable deflection and stress changes were observed between the web-
105
plumbness and web out-of-plumbness cases. Therefore, the effects of web-plumbness on curved and skewed bridge
constructability are considered not significant.
In summary, findings from the web-plumbness parametric studies for the curved and skewed bridges that were
examined included:
Curved
o Web out-of-plumbness did not cause appreciable bridge deflection and stress increases when the
out-of-plumbness was within the limit (1%) specified in the Structural Welding Code (AWS,
2004).
o Exceeding the 1% limit of the web out-of-plumbness can result in slightly higher vertical and
lateral deformations and also stresses. However, the effects of horizontal curvature on these
parameters are much larger than those from the web out-of-plumbness.
Skewed
2004).
o Exceeding the 1% limit of the web out-of-plumbness can result in slightly higher vertical
deformations and stresses. However, the effects on lateral deformations are more pronounced. As
a result, the effects of web-out-of plumbness, which could be beneficial to ensure web-plumb at
the completion of construction, should be considered during the erection of structures having a
skew less than 70 .
5.4
Temporary Shoring Placement and Settlement Effects
5.4.1
5.4.1.1
Parametric Studies
Curved
Eight representative bridges were selected to investigate the effects of temporary shoring placement on curved
bridge behavior during construction. The selected bridges were two-span structures with varying radii and crossframe spacings and three-span structures with balanced and unbalanced spans. The radii of the two-span bridges
were 91.4 m (300 ft), 198.1 m (650 ft), and 304.8 m (1000 ft), and the cross-frame spacings were 4.57 m (15 ft) and
6.86 m (22.5 ft). These parameters were selected so that they included moderately to severely curved bridges with
different cross-frame spacings, resulting in R/L values from 13.3 to 66.7. The balanced and unbalanced three-span,
four-girder bridges had the same radius (91.4 m (300ft)) and cross-frame spacing (6.86 m (22.5 ft)), with the
unbalanced span ratio being 1:1.4. Sequential analysis was performed to simulate bridge behavior for each
construction stage. The paired inner erection method was adopted in this study since it was considered the preferred
erection approach from the erection sequencing investigations.
106
Table 25 lists the parameters of the selected representative bridges.
107
Bridge
No.
C1
C3
C6
C7
C9
C10
C11
C12
Table 25: Selected Curved Bridge Temporary Shoring Study Information.
CrossRadius of
GirderFrame
Number
Curvature, m
Spacing,
Span-Length, m (ft)
Spacing, m
of Spans
(ft)
m (ft)
(ft)
2
68.6-68.6
91.4 (300)
4.57 (15)
3 (10)
(225-225)
2
68.6-68.6
91.4 (300)
6.86 (22.5)
3 (10)
(225-225)
2
68.6-68.6
198.1 (650)
6.86 (22.5)
3 (10)
(225-225)
2
68.6-68.6
304.8 (1000)
4.57 (15)
3 (10)
(225-225)
2
68.6-68.6
304.8 (1000)
6.86 (22.5)
3 (10)
(225-225)
3
68.6-68.6-68.6
91.4 (300)
6.86 (22.5)
3 (10)
(225-225-225)
3
48-68.6-68.6
91.4 (300)
6.86 (22.5)
3 (10)
(157.5-225-225)
3
48-68.6-48
91.4 (300)
6.86 (22.5)
3 (10)
(157.5-225-157.5)
Number
of
Girder
4
4
4
4
4
4
4
4
Three shoring conditions were investigated to determine their effects on bridge constructability. The first shoring
condition was to place supports at splice locations only; the second condition was to place supports at splice
locations as well as at 0.4L (for the exterior span) and 0.5L (for the interior span) of the span when the span was
longer than 100 ft; and the third condition was to place supports at 0.75L of the span.
Figure 145 through Figure 148 show the three shoring conditions for the selected two-span and three-span bridges.
108
a)
Shoring condition 1
b) Shoring condition 2
c)
Shoring condition - 3
Figure 145: Shoring Conditions for Two-Span Bridges.
109
a)
Shoring condition 1
Figure 146: Shoring Conditions for Bridge C10.
a)
Shoring condition-1
110
a)
Shoring condition-1
b) Shoring condition-2
After the study of temporary shoring effects, the effects of settlement were studied for Bridge C3, which was the
most severely curved structure with the largest cross-frame spacing. A total of six settlement conditions were
examined (listed in Table 26), which included two settlement scenarios having three different settlement
magnitudes. The first scenario introduced a settlement at one support at one splice location, while the second
scenario introduced a settlement at supports at two splice locations. The three different settlement magnitudes that
were applied were 0.001L, 0.002L, and 0.004L, where L is the span length between the abutment and pier. These
111
values gave settlements between approximately 2 ½"and 10". Consequently, the effects of increasing settlement
could be observed.
Settlement conditions
Condition-1
Condition-2
Condition-3
Condition-4
Condition-5
Condition-6
5.4.1.2
Table 26: Selected Settlement Conditions.
Scenario
Settlement at one support
Settlement at two supports
Settlement
0.001L
0.002L
0.004L
0.001L
0.002L
0.004L
Skewed
Eight representative bridges were also selected from the skewed bridge set to investigate the effects of temporary
shoring placement on bridge behavior during construction. The selected bridges included one single-span bridge
with severely skewed abutments and maximum cross-frame spacing, four two-span bridges with varying skew
angles and cross-frame spacings and three three-span bridges with balanced and unbalanced spans. The skew angle
of the single-span bridge was 50 , and the cross-frame spacing was 7.8 m (25.7 ft). The varying skew angles of the
two-span bridges were 50 and 70 , each with two different cross-frame spacings of 4.57 m (15 ft) and 7.8 m (25.7
ft). These parameters were selected so that they included a range from moderately to severely skewed bridges with
different cross-frame spacings. Two symmetric three-span bridges with balanced and unbalanced spans and one nonsymmetric three-span bridge with unbalanced spans were also included in this parametric study. All three-span
bridges had the same skew angle (50 ), the same cross frame spacing (7.8 m (25.7 ft)), and a 1:1.4 span ratio.
Similar to the curved bridges, the paired-girder erection method was adopted for analysis consistency between
curved and skewed models. Table 27 lists the parameters of the selected representative bridges.
Bridge
No.
S2
S5
S6
S7
S8
S9
S10
S11
Table 27: Selected Skewed Bridge Temporary Shoring Study Information.
CrossNumber
Skew
GirderNumber
Frame
of
Angle,
Spacing, m
Span-Length, m (ft)
of
Spacing,
Spans
degrees
(ft)
Girder
m (ft)
3 (10)
50
7.8 (25.7)
1
54.9 (180)
4
3 (10)
54.9-54.9
50
4.57 (15)
2
4
(180-180)
3 (10)
54.9-54.9
50
7.8 (25.7)
2
4
(180-180)
3 (10)
54.9-54.9
70
4.57 (15)
2
4
(180-180)
3 (10)
54.9-54.9
70
7.8 (25.7)
2
4
(180-180)
3 (10)
54.9-54.9-54.9
50
7.8 (25.7)
3
4
(180-180-180)
3 (10)
54.9-54.9-39.2
50
7.8 (25.7)
3
4
(180-180-128.75)
3 (10)
39.2-54.9-39.2
50
7.8 (25.7)
3
4
(128.75-180-128.75)
Similar to the study of the representative curved bridges, three shoring conditions were investigated to determine
their effects on bridge constructability. The first shoring condition was to place supports at splice locations only, the
second condition was to place supports at splice locations as well as at 0.4L of the span for exterior spans and and/or
0.5L for interior spans, and the third condition was to place supports at 0.75L of the first span and at splice locations
for the second span of the two-span bridges. Figure 149 through Figure 151 show the three shoring conditions for
the selected two-span and three-span bridges.
112
a)
Shoring condition-1
Figure 149: Shoring Conditions for Single-Span Skewed Bridge.
a)
Shoring condition-1
c)
Shoring condition-3
Figure 150: Shoring Conditions for Two-Span Skewed Bridges.
113
a)
Shoring condition-1
Figure 151: Shoring Conditions for Three-Span Skewed Bridges.
Similar to the study of the representative curved bridges, the settlement effects on the skewed structures were
investigated by considering six settlement conditions (Table 26) for Bridge S6 (the most severely skewed structure
with the largest cross-frame spacing). The first settlement scenario introduced a settlement at one support at one
splice location, while the second scenario introduced settlement at supports at both splice locations. The three
different settlements applied were 0.001L, 0.002L, and 0.004L, where L is the span length between the abutment
and pier, which gave settlement values between approximately 2‖ and 9‖. Consequently, the effects of increasing
settlement could be observed.
5.4.2
5.4.2.1
Curved
Maximum girder vertical deflections and Von Mises stresses in cross frames were compared to study the effects of
temporary shoring placement on bridge behavior during construction. Figure 152 through Figure 159 show the
comparisons of deflections between the first two shoring conditions. Not surprisingly, adding additional supports in
girders with span lengths of more than 100 ft considerably reduced girder deflections, leading to a more stable
structure and therefore a more desirable construction condition. Secondly, considering that under certain
circumstances contractors might not be able to provide the additional shoring between supports, the third shoring
condition (shoring at 0.75L) was studied to provide an alternative to contractors for mitigating girder deflections.
Figure 160 through Figure 165 show the representative results of vertical deflections and Von Mises stresses for all
three shoring conditions. Interestingly, as observed from the results, the maximum girder deflections and stresses in
cross frames could be significantly reduced by placing one shoring tower at 0.75L. The results also show that more
reductions in deflections and stresses were observed for severely curved structures than for moderately curved
structures when a temporary support was provided at 0.75L instead of at splice locations. This finding suggested
that, for the erection of a long span (>100 ft), severely curved (R=300 ft) bridges, placing one support between the
splice location and mid-span could increase the stability and constructability of the structure over one that does not
have any intermediate shoring.
114
115
116
117
118
119
Figure 163: Ratio of Maximum Von Mises Stresses for Bridge C3.
120
Settlement effects on the stresses in cross frames were also examined for the six settlement conditions. The cross
frame having the maximum Von Mises stress before support settlement was checked for stress increases after
settlement.
Figure 166 shows the results for the ratio of maximum Von Mises stresses. First, as seen in
Figure 166, stresses were increased due to the support settlements; nevertheless, no significant difference in the
increase of stress was observed between the one-support settlement and two-support settlement conditions, with the
two-support settlement condition having a slightly higher stress increase. Second, stresses were increased
proportionally with increasing settlement. An average of 25% stress increase in cross frames was observed with
every increase of 0.0001L settlement for the bridge studied.
121
Figure 166: Ratio of Maximum Von Mises Stress for Bridge C3 with Various Settlement Conditions.
5.4.2.2
Skewed
Maximum vertical deflections in the girders and maximum Von Mises stresses in the cross frames were also
examined for the skewed bridge shoring cases. Figure 167 through Figure 174 show the comparisons of deflections
between the first two shoring conditions for all representative bridges and all construction stages. As was expected,
additional shoring at the mid spans of the girders dramatically decreased the maximum vertical deformations which,
in turn, could enhance the stability of all the bridges. The behavior of the structures under the third shoring condition
(at 0.75L) was studied by looking at the results from vertical deflections for the two-span representative skewed
bridges at the second stage of construction. The results for the other stages were shown to be the same as the the
results for Shoring condition-1 because of the similarity of the support conditions for those stages. Figure 175
through Figure 178 show the ratios of maximum girder vertical deformation at Stage-2. Finally, maximum Von
Mises stresses in the cross frames were then compared in the two-span bridges for all three shoring conditions, and
results are shown in Figure 179 through Figure 182. Similar to the representative curved bridges, in comparison with
Shoring condition-1 (placing the shoring at a splice location), the maximum girder deflections and stresses in cross
frames could be significantly reduced by moving the shoring to 0.75L of the first span. For all three conditions, the
ratios of maximum deflection and maximum Von Mises stress are nearly the same for all four examined skewed
bridges.
122
123
124
125
126
Figure 175: Ratio of Maximum Vertical Deflections for Three Shoring Conditions in Bridge S5.
Figure 176: Ratio of Maximum Vertical Deflections for Three Shoring Conditions in Bridge S-6.
127
128
Figure 179: Ratio of Maximum Von Mises Stresses for Bridge S5.
129
The effect of settlement on cross-frames stresses was also examined for the six settlement conditions. The cross
frame with the maximum Von Mises stress before support settlement was checked for stress increases after
settlement. Figure 183 shows the results for the ratio of maximum Von Mises stresses. In comparison with the
results from the curved bridges, the cross frames experienced a relatively large increase in stress for all settlement
conditions. It can also be observed from Figure 183 that, unlike the curve bridges, there is an appreciable difference
between the effects of settlement of a single support and settlement of two supports for skewed bridges. The reason
for these effects is the relatively small stresses on the cross frames due to dead load torsional effects in the skewed
bridges for the no settlement condition. By comparison, in the curved bridges, the cross frames experienced larger
130
stresses as a result of curvature effects. Therefore, the increase in stress for skewed bridges becomes more prominent
because it largely reflects the settlement effects. However, the magnitudes of these stresses are smaller overall than
those in curved bridges under settlement effects.
Figure 183: Ratio of Maximum Von Mises Stress for Bridge S6 with Various Settlement Conditions.
5.4.2.3
Summary
This section examined temporary shoring and settlement effects on curved and skewed bridge behavior during
construction. In the first part of this study, the effectiveness of three shoring scenarios (1. splice locations; 2. splice
locations and 0.4L; 3. 0.75L) was investigated. It was observed that placing one temporary shoring at 0.75L between
the abutment and splice location effectively reduced girder deflections and improved bridge constructability for both
curved and skewed structures. The second part of this study investigated settlement effects on curved and skewed
bridge behaviors during construction. Two settlement scenarios (one-support settling and two-supports settling) with
three various amounts of settlement were studied. Results showed appreciable cross-frame stress increases due to
support settlement effects. Therefore, settlement effects on bridge constructability should be taken into account for
construction planning.
In summary, findings from the temporary shoring parametric studies for the curved and skewed bridges that were
examined included:
Curved
o In addition to the shoring at splice locations, placing one temporary support at 0.4L could
significantly reduce girder deflections, leading to a more constructible condition.
o
When adding additional shoring at 0.4L was not feasible, placing one support at 0.75L between
the abutment and splice location could reduce girder vertical deflections by more than 75% for the
structures studied when compared to the deflections of placing one support at the splice location.
o
Attempts should be made to limit support settlements to less than 0.001L (L=span length) during
construction to mitigate stress increases in bridge members.
o
Small differences were observed between one-support settling and two-supports settling.
131
Skewed
5.5
o
In addition to the shoring at splice locations, placing one temporary support at 0.4L could
significantly reduce girder deflections, leading to a more constructible condition.
o
o
Attempts should be made to limit support settlement to less than 0.001L (L=span length) during
o
Two-supports settling had a larger impact on stresses than one-support settling.
Cross Frame Consistent Detailing
Earlier work (Chavel and Earls 2006) indicated that detailing curved girders and cross frames for different load
conditions during construction can lead to undesirable and costly fit-up problems. Typically, this inconsistent
detailing occurs when girders are detailed to be web-plumb under one construction condition (e.g., no load) and
cross frames are detailed to be web-plumb under another condition (e.g., full dead load). To model these effects in
ABAQUS, the studied curved and skewed bridge models were initially run in SAP 2000 for a web-plumb under no
load condition to define the geometry of cross frames and girders in the ABAQUS models. The SAP models also
assisted with finding the geometry of inconsistently detailed cross frames by providing vertical deformations of the
girders at their connection with each cross frame under full dead load. The new geometries of the cross frames were
defined assuming the webs were plumb in this deformed shape, as shown in Figure 184. When compared against a
consistent detailing case, the geometries resulted in elongation of one cross-frame diagonal and shortening of the
other. To mimic the forces applied to the structures as a result of this inconsistent detailing, positive or negative
temperatures that would give these length changes were calculated for each diagonal. These temperatures were then
applied the members in the ABAQUS models to investigate the effects of inconsistent detailing on curved and
skewed bridge constructability.
a)
5.5.1
5.5.1.1
Deformed shape due to full dead load from
b) Deformed shape, assuming web-plumb
SAP 2000 models (web-plumb under no
under full dead load condition
load condition)
Figure 184: Cross Frame Geometries for Inconsistent Detailing.
Parametric Studies
Curved
To examine inconsistent detailing effects on curved bridges, one two-span curved bridge with a severe curvature and
one unbalanced three-span bridge were selected.
132
Table 28 lists parameters for the bridges that were examined.
133
Bridge
No.
C3
C11
5.5.1.2
Table 28: Selected Inconsistent Detailing Study Bridge Information
CrossRadius of
GirderFrame
Number
Curvature, m
Spacing,
Span-Length, m (ft)
Spacing, m
of Spans
(ft)
m (ft)
(ft)
2
68.6-68.6
91.4 (300)
6.86 (22.5)
3 (10)
(225-225)
3
48-68.6-68.6
91.4 (300)
6.86 (22.5)
3 (10)
(157.5-225-225)
Number
of
Girder
4
4
Skewed
Similar to the curved bridges, the effects of the inconsistent detailing of the cross frame on the constructability of the
skewed bridges was investigated by considering four two-span bridges with varying skew angles and cross-frame
spacings. Table 29 lists the parameters for the bridges that were examined.
Table 29: Selected Skewed Bridge Consistent Detailing Study Bridge Information
CrossNumber
Skew
Number
Bridge
Frame
Girderof
Angle,
Span-Length, ft
of Girder,
No.
Spacing,
Spacing, ft
Spans
degree
ft
ft
10
S5
50
15
2
180-180
4
10
S6
50
25.7
2
180-180
4
10
S7
70
15
2
180-180
4
10
S8
70
25.7
2
180-180
4
5.5.2
5.5.2.1
Curved
To examine the effects of inconsistent detailing on bridge constructability and the resulting locked-in stresses in the
cross frames, girder splice vertical and radial deflections and cross-frame Von Mises stresses were again examined.
Comparisons are shown in Figure 185 through Figure 190. As shown in the results, inconsistent cross-frame
detailing resulted in higher vertical and radial deflections. More importantly, it was also observed that stresses in
cross frames were increased due to locked-in stresses caused by inconsistent detailing. This change in stress
occurred more or less in all cross frames along the bridge; however, cross frames near the supports experienced
slightly higher increases.
134
135
136
5.5.2.2
Skewed
Similar to the representative curved bridges, to examine the effects of inconsistent detailing on bridge
constructability and the resulting locked-in stresses in the skewed bridge cross frames, girder splice deflections and
cross-frame Von Mises stresses were examined. Results are shown in Figure 191 through Figure 202. The results
show a very small change in vertical deflection in all representative bridges for consistent and inconsistent detailing.
However, inconsistent detailing appreciably increased girder lateral deflections and Von Mises stresses in the cross
frames. In similar fashion to the curved bridges, cross frames near the supports experienced higher stresses as a
result of inconsistent detailing. Moreover, increases for bridges with larger cross-frame spacings were slightly larger
than those with smaller cross-frame spacings.
137
138
139
140
141
142
5.5.2.3
Summary
This section examined the effects of inconsistent cross-frame detailing on bridge behavior during construction for
curved and skewed structures. Temperature loads were applied to cross frames to mimic the external forces applied
to them during construction. The inconsistent detailing was caused by detailing girders and cross frames for different
load conditions. In this study, girders were detailed to be web-plumb under no load, and cross frames were detailed
to be web-plumb under full dead load. The results suggest that, for both curved and skewed structures, inconsistent
cross-frame detailing had a detrimental impact on girder rotations and cross-frame stresses, but had little influence
on girder vertical deflections.
Curved
o Inconsistent cross-frame detailing increased vertical and radial deflections.
o Cross-frame stresses were appreciably increased as a result of the locked-in stresses due to the
inconsistent detailing. This increase was more pronounced for cross frames near the supports.
o Irrespective of the radius and cross-frame spacing, inconsistent cross-frame detailing should be
avoided for all curved structures to mitigate locked-in stresses and increased deformations. The
143
effects of inconsistent detailing can be more pronounced in bridges with larger cross frame
spacings.
Skewed
o Inconsistent cross-frame detailing did not have a crucial impact on girder vertical deflections.
o Inconsistent cross-frame detailing increased lateral deflections.
o Cross-frame stresses were appreciably increased as a result of the locked-in stresses due to
o Irrespective of the skew and cross-frame spacing, inconsistent cross-frame detailing should be
avoided for all skewed structures to mitigate locked-in stresses and increased deformations. The
spacings.
5.6
Solid Plate Diaphragms
This section explored the effects of replacing cross-frame members with solid plate diaphragms on the construction
response of horizontally curved and skewed steel bridges. Three scenarios to replace cross frames with diaphragms
were investigated: (1) diaphragms at abutments only, (2) diaphragms at abutments and piers, and (3) diaphragms at
abutments, piers and mid-span locations. Diaphragms were designed as short, deep plate girder sections and were
initially sized according to appropriate slenderness limits from AASHTO and PennDOT DM4. Once again, SAP
2000 models were employed for final design where appropriate cross frames were replaced. Design forces were then
extracted from the analysis results, and the initial cross sections were checked for bending and shear based on
relevant AASHTO and Penn DOT DM4 criteria. These final diaphragm sections were used in the subsequent
ABAQUS models in place of the cross frames. In these models, similar to the study performed for the girders,
diaphragms were modeled utilizing beam elements for their top and bottom flanges and shell elements for their webs,
and were tied to the girder webs.
5.6.1
5.6.1.1
Parametric Studies
Curved bridges
Cross frames and diaphragms in curved bridges play an important role in transferring loads, distributing live and
dead loads to the girders, providing stability to the compression flanges during erection, and providing torsional
stiffness for the overall structure. The examination of the effects of replacing cross frames with diaphragms at
abutments, piers, and mid-span locations focused on four representative bridges. The radii of the selected curved
bridges were 91.4 m (300 ft) and 304.8 m (1000 ft), and the cross-frame spacings were 4.57 m (15 ft) and 6.86 m
(22.5 ft). Table 30 lists the parameters for the bridges that were examined. The paired inner erection method was
used for all sequential analyses.
Bridge
No.
C1
C3
C7
C9
Table 30: Selected Curved Bridge Diaphragm Study Bridge Information.
CrossRadius of
GirderFrame
Number
Curvature, m
Spacing,
Span-Length, m (ft)
Spacing, m
of Spans
(ft)
m (ft)
(ft)
68.6-68.6
91.4 (300)
4.57 (15)
3 (10)
2
(225-225)
68.6-68.6
91.4 (300)
6.86 (22.5)
3 (10)
2
(225-225)
68.6-68.6
304.8 (1000)
4.57 (15)
3 (10)
2
(225-225)
68.6-68.6
304.8 (1000)
6.86 (22.5)
3 (10)
2
(225-225)
144
Number
of
Girder
4
4
4
4
5.6.1.2
Skewed bridges
Similar to their role in curved bridges, cross frames and diaphragms in skewed bridges with skew angles smaller
than 70 (Penn DOT, 2007) play an important role in load distribution and enhancing stability of the overall
structure. As for the curved bridges, the effects of replacing cross frames with diaphragms at sections close to the
abutments, piers, and mid-span locations on skewed bridge construction response were studied by looking at four
representative two-span bridges. The selected bridges had skew angles of 50 and 70 and cross-frame spacings of
4.57 m (15 ft) and 7.8 m (25.7 ft). Table 31 lists the parameters for the skewed bridges that were examined.
Table 31: Selected Curved Bridge Diaphragm Study Bridge Information.
CrossSkew
GirderNumber
Number
Bridge
Frame
Angle,
Spacing, m
of
Span-Length, m (ft)
of
No.
Spacing,
degrees
(ft)
Spans
Girder
m (ft)
54.9-54.9
S5
50
4.57 (15)
3 (10)
2
4
(180-180)
54.9-54.9
S6
50
7.8 (25.7)
3 (10)
2
4
(180-180)
54.9-54.9
S7
70
4.57 (15)
3 (10)
2
4
(180-180)
54.9-54.9
S8
70
7.8 (25.7)
3 (10)
2
4
(180-180)
5.6.2
5.6.2.1
Curved bridges
At each stage the maximum girder vertical deflection and maximum radial deflections at splice locations were
examined and normalized against maximum values of the base model, which was the bridge with all cross frames
(no diaphragms). Furthermore, maximum Von Mises stresses were compared in cross-frame members after
completed erection of the superstructure and deck placement for each scenario. Figure 203 through Figure 214 show
comparisons of the deflections and stresses.
Results indicated that the use of diaphragms instead of cross frames caused an increase in both vertical and radial
deflections and Von Mises stresses. Diaphragms replacing cross frames at abutments only or at abutments and pier
(scenario 1 and 2) showed an average 5% increase in maximum girder vertical deflection. However, when the cross
frames were also replaced with diaphragms at mid-span locations (scenario 3), it caused a vertical deflection
increase of 7% in 4.57m (15 ft) cross-frame spacing and 9% in 6.86m (22.5 ft) cross-frame spacing, compared to
bridge models without diaphragms. Von Mises stresses in cross frames showed an average increase of 13% in all
models when compared to the base models with no diaphragms. Diaphragms placement for all 3 scenarios caused an
average increase of 10% in radial deflections at the first splice location, and no major changes in deflections at the
second splice location. Consequently, it is concluded from these results that the addition of diaphragms instead of
cross frames at support and mid-span locations has no appreciable effects on reducing deflections and stresses
during the construction of horizontally curved steel bridges.
145
146
147
148
149
150
5.6.2.2
Skewed bridges
Vertical deflection of the girders, lateral deformations at splice locations, and maximum Von Mises stresses in cross
frames were also compared for the representative skewed bridges. These values were normalized with their
151
corresponding values for the bridge models which had only cross frames and no diaphragms. Figure 215 through
Figure 226 show comparisons of the deformations and stresses.
As can be observed in the results, maximum vertical deformation of the girders at mid-span and lateral deformations
at splice location increased slightly for bridges with diaphragms when compared to those values for bridges with no
diaphragms. The reason for this increase is the additional load imposed to the structure due to the self weight of the
diaphragm elements, which are relatively heavy when compared to the cross frames they replaced. This increase in
deformations is higher, to some degree, for those bridges that have larger cross-frame spacing (Bridge S-6 and
Bridge S-8). For maximum stresses in cross frames, the effect of replacing cross frames with diaphragms is not
significant for any of the cases. In general, it can be concluded from these results that using diaphragms instead of
cross frames at supports and mid-span locations of the skewed bridges had insignificant influence on their behavior
during construction.
152
153
Figure 219: Ratio of Maximum Lateral Deflections for Bridge S5.
154
155
Figure 223: Ratio of Maximum Von Mises Stresses in Cross Frames for Bridge S5
156
157
5.6.2.3
Summary
This section examined the effects of replacing cross frames with diaphragms at the abutments, piers, and mid-span
locations on curved and skewed bridge behavior during construction. Results showed no significant changes in
girder deflections and cross-frame stresses between bridges with diaphragms and without diaphragms. The girder
deflections and cross-frame stresses were slightly increased due to the diaphragm self weight. Therefore, replacing
cross frames with diaphragms had no appreciable benefit to the bridge constructability for both curved and skewed
structures.
In summary, findings from the diaphragm parametric studies for the curved and skewed bridges that were examined
included:
Curved
o Replacing cross frames with solid plate diaphragms did not severely affect bridge vertical and
radial deflections and cross-frame stresses during construction, irrespective of the radius and
cross-frame spacing.
o Replacing cross frames with solid plate diaphragms at the abutment and pier locations for the
curved bridges that were studied did not adversely affect or appreciably benefit their construction
behavior. Nevertheless, using diaphragms at the mid-span locations can cause slightly higher
deformations for the girders when compared to the use of cross frames at those locations.
Skewed
o Placing solid plate diaphragms in skewed bridges slightly increased deformations due to the self
weight of the diaphragms, but did not severely affect cross-frame stresses.
skewed bridges did not adversely affect or appreciably benefit their construction behavior.
However, using diaphragms at the mid-span locations can cause slightly higher deformations for
the girders when compared to the use of cross frames at those locations.
5.7
Temperature Change
This section examined the effects of temperature change during deck placement. To study this effect, daily
temperature change data from National Oceanic and Atmospheric Administration’s (NOAA) National Weather
Service (2010) for central Pennsylvania was used to determine the value of temperature change to be applied to
158
structures during a given construction event. Figure 227 presents daily high and low temperatures in central
Pennsylvania from 2003 to 2007 obtained from NOAA. The maximum temperature change in a day from 2003 to
2007 was 43 °F. To be conservative, a 50 °F temperature rise was applied to the structures during deck placement,
so the resulting maximum cross-frame stress was the highest. To model temperature effects in ABAQUS,
temperature loads were applied to the entire steel superstructure, with the initial temperature being 20 °F and the
final temperature being 70 °F. These values were arbitrarily selected to give the desired temperature change, and
their actual magnitudes had no bearing on ABAQUS results. Deflections and stresses were monitored before and
after the temperature change.
Figure 227: Central Pennsylvania High and Low Temperatures. (NOAA, 2010).
5.7.1
5.7.1.1
Parametric Studies
Curved
Examination of the effects of temperature change during deck placement on curved structures consisted of
comparisons of deflection and stress changes after the temperature change occurred. To investigate the effects of
radius, cross-frame spacing, and unbalanced spans in conjunction with temperature change during deck placement,
eight representative bridges involving different parameters (radius, cross-frame spacing, and span ratio) were
selected. The selected bridges were two-span structures with varying radii and cross-frame spacings and three-span
structures with balanced and unbalanced spans. The radii of the two-span bridges were 91.4 m (300 ft), 198.1 m
(650 ft), and 304.8 m (1000 ft), and the cross-frame spacings were 4.57 m (15 ft) and 6.86 m (22.5 ft). These
parameters were selected so that they included moderately to severely curved bridges with different cross-frame
spacings, resulting in R/L values from 13.3 to 66.7. The balanced and unbalanced three-span, four-girder bridges
had the same radius (91.4 m (300ft)) and cross-frame spacing (6.86 m (22.5 ft)), with the unbalanced span ratio
being 1:1.4. Table 32 lists the parameters of the selected representative bridges.
Bridge
No.
C1
Table 32: Selected Curved Bridge Temperature Change Study Information.
CrossRadius of
GirderNumber
Frame
Number
Curvature, m
Spacing,
Span-Length, m (ft)
of
Spacing, m
of Spans
(ft)
m (ft)
Girder
(ft)
2
68.6-68.6
91.4 (300)
4.57 (15)
3 (10)
4
(225-225)
159
5.7.1.2
C3
91.4 (300)
6.86 (22.5)
3 (10)
C6
198.1 (650)
6.86 (22.5)
3 (10)
C7
304.8 (1000)
4.57 (15)
3 (10)
C9
304.8 (1000)
6.86 (22.5)
3 (10)
C10
91.4 (300)
6.86 (22.5)
3 (10)
C11
91.4 (300)
6.86 (22.5)
3 (10)
C12
91.4 (300)
6.86 (22.5)
3 (10)
2
2
2
2
3
3
3
68.6-68.6
(225-225)
68.6-68.6
(225-225)
68.6-68.6
(225-225)
68.6-68.6
(225-225)
68.6-68.6-68.6
(225-225-225)
48-68.6-68.6
(157.5-225-225)
48-68.6-48
(157.5-225-157.5)
4
4
4
4
4
4
4
Skewed
Similar to the curved bridges, the effects of temperature change during deck placement was investigated by
considering eight representative bridges. These bridges were selected to include the effects of skew angle, crossframe spacing, and unbalanced spans in conjunction with temperature change. The selected bridges were two singlespan bridges with severely skewed abutments (50 ) and varying cross-frame spacing, four two-span bridges with
varying skew angle (50 and 70 ) and cross-frame spacing, and two three-span bridges with balanced and
unbalanced spans. The cross frame-spacings in the single- and two-span bridges were 4.57 m (15 ft) and 7.8 m
(25.7 ft). The balanced and unbalanced three-span, four-girder bridges had the same skew angle (50 ) and crossframe spacing (7.8 m (25.7 ft)), with the unbalanced span ratio being 1:1.4. Table 33 lists the parameters of the
selected representative bridges.
Table 33: Selected Skewed Bridge Temperature Change Study Information.
CrossNumber
Skew
GirderNumber
Bridge
Frame
of
Angle,
Spacing, m
Span-Length, m (ft)
of
No.
Spacing,
Spans
degrees
(ft)
Girder
m (ft)
3 (10)
S1
50
4.57 (15)
1
54.9 (180)
4
3
(10)
S2
50
7.8 (25.7)
1
54.9 (180)
4
3 (10)
54.9-54.9
S5
50
4.57 (15)
2
4
(180-180)
3 (10)
54.9-54.9
S6
50
7.8 (25.7)
2
4
(180-180)
3 (10)
54.9-54.9
S7
70
4.57 (15)
2
4
(180-180)
3 (10)
54.9-54.9
S8
70
7.8 (25.7)
2
4
(180-180)
3 (10)
54.9-54.9-54.9
S9
50
7.8 (25.7)
3
4
(180-180-180)
3 (10)
54.9-54.9-39.2
S10
50
7.8 (25.7)
3
4
(180-180-128.75)
5.7.2
5.7.2.1
Curved
Comparisons of deflections and stresses are presented in Figure 228 through Figure 2. Girder deflection changes
between bridges studied caused by the temperature change were not as noticeable as stress changes in the cross
frames. The results showed that stress changes due to temperature increased with increasing curvature and
160
decreasing cross-frame spacing. These findings suggested: (1) temperature changes could aggravate curvature
effects on cross-frame stresses, and (2) smaller cross-frame spacings could lead to higher stress increases caused by
temperature changes.
161
162
163
164
165
166
167
5.7.2.2
Skewed
Ratios of maximum girder vertical deflections and cross-frame stresses are compared in Figure 244 through Figure .
There is no marked change in vertical deformation or stresses due to temprature change in skewed bridges.
168
169
170
171
172
173
174
175
5.7.2.3
Summary
This section examined the effects of temperature changes on curved and skewed bridge constructability during deck
placement. In the analysis, temperature was increased by 50 °F after deck placement. Girder deflections and crossframe stresses were examined after the temperature rise. Results showed minor changes in bridge deflections and
stresses due to the temperature change. Hence, it was considered that temperature changes during construction had
little impact on bridge constructability.
In summary, findings from the temperature change parametric studies for the curved and skewed bridges that were
examined included:
Curved
o The applied temperature change did not have an appreciable impact on overall bridge deflections
and stresses for all of the radii and cross-frame spacings studied.
176
Skewed
and stresses for all of the skew angles and cross-frame spacings studied.
6
6.1
CONCLUSIONS
Project Overview
Summarized herein were research activities and findings in association with Work Order 009, ―Guidelines for
Analyzing Curved and Skewed Bridges and Designing Them for Construction.‖ As stated in the introduction, the
project was divided into seven tasks:
1. Updated Literature Search
2. Data Acquisition System Maintenance
3. Numerical Modeling
4. Parametric Studies
5. Draft Final Report
6. Final Report
7. Invoice Submission
This report focused on Tasks 1 through 5, and due to earlier submittals related to Tasks 1 through 3 (Hiltunen et al.
2004; Linzell et al. 2003; Linzell et al. 2006; Linzell et al. 2008; Linzell et al. 2008), the bulk of this document
summarizes the work completed and findings developed for Task 4. However, some information related to Task 2 is
also provided.
The findings and the numerical modeling from the parametric studies are the basis for suggesting possible
modifications to relevant PennDOT publications. Those publications are included in Appendix D of this report with
suggested changes being shown in italics.
In-service structural response was tracked using instruments discussed in previous submittals to PennDOT. Using
instruments that were in place on two bridges (Structure #’s 207 and 314), responses were tracked due to
temperature changes after construction was completed. Data provided by these instruments for the examined time
period did not indicate any obvious trends with respect to their global behavior.
The parametric studies that were completed focused on a group of representative curved and skewed steel bridge
structures and numerically examined the influence of specific variables, as identified in the project scope, on
behavior during construction. These items included: (1) web-plumbness; (2) temporary shoring placement and
settlement effects; (3) cross-frame consistent detailing (i.e., applying the work by Chavel and Earls to other bridge
geometries); (4) girder and cross-frame erection sequencing along the span and with respect to girder radius and the
effects of ―drop-in‖ erection; (5) solid plate diaphragms verses cross frames; and (6) global temperature change
effects. Finite element models assisted with preliminary and final designs and with final, sequential analyses that
focused on the influence of the above items on deformations and stress states at the completion of construction.
Design and preliminary models were developed in SAP2000, with the final models used for the sequential analyses
being developed in ABAQUS. Modeling decisions and approaches are summarized in previous submittals to
PennDOT (Linzell et al. 2008). A superstructure modeling approach that consisted of representing the girder webs
and concrete deck using shell elements, with beam elements representing other major components, such as the girder
flanges and cross frames, was used. This modeling technique provided an acceptable compromise between reduced
computation times provided by grillage models and increased accuracy provided by more sophisticated, and
complicated, three-dimensional finite element models.
The final parametric studies were completed for a specified set of curved and skewed steel bridges, with bridge
global geometries being obtained from a combination of statistical studies and PennDOT input. A total of 12 final
curved bridge designs and 11 final skewed bridge designs were developed for the parametric studies. Information
related to the final designs can be found in Table 3 and Table 4. In addition to these final structure studies, initial
parametric studies that investigated the effects of erection sequencing on an additional, larger group of curved
177
bridges were completed prior to studying erection sequencing effects on the final curved bridge design group.
Justifications for completing these initial studies and the bridges that were examined are discussed below.
6.2
Parametric Study Findings
Erection sequencing was the first parameter numerically studied using ABAQUS because the findings from that
portion of the project influenced the remaining items that were examined. The study of the effects of erection
sequencing began with initial examinations of a broad range of single- and two-span curved, I-girder structures and
a single three-span, curved, I-girder structure. Various erection sequencing scenarios used in the initial studies were
then applied to the final 12 curved bridges shown in Table 3 to reaffirm findings from the initial studies. Initial
studies were not completed for the skewed structures, and erection sequencing scenarios recommended for the final
curved structures were applied to the skewed structures in Table 4 to study their effects on construction response.
The initial studies examined a total of thirty single- and two-span bridges and a single three-span structure under the
influence of the following erection sequencing parameters: (1) varying radii; (2) single-span structures and two-span
structures with varying span ratios; (3) 4- and 5-girder cross sections; (4) different erection sequencing options that
including erecting single girders and girders in pairs; (5) erecting the girders from inner to outer radius of curvature
and from out to inner radius of curvature; and (6) the influence of temporary shoring. A summary of the initially
examined single- and two-span bridges can be found in Figure 28. A framing plan of the single three-span structure
that was initially examined is shown in Figure 59. These structures were statically analyzed in ABAQUS for the
following construction scenarios: (1) paired-girder erection placing the interior girders first; (2) paired-girder
erection placing the exterior girders first; (3) single-girder erection that placed the interior girder first; and (4) singlegirder erection that placed the exterior girder first. ―Paired‖ girder erection for the 5-girder bridges involved placing
a single girder at some point during the process. Results from the initial group of analyses were examined
statistically to identify preferred erection sequencing approaches, with preferred sequences being identified
predominantly based on deflections, with vertical and radial deflections being of primary importance. Results from
the initial studies indicated that, for the structures examined, the construction methods that initiated with the inner
(lowest radius) girder are preferred, irrespective of number of spans, and that for bridges having more than two
spans, paired inner erection methods are preferred. Results also indicated that, irrespective of span number and
geometry, construction that initiates with a single outer girder is not recommended. In addition, it was observed
from the results that bridge deflections were mainly controlled by R/L values, boundary conditions, and span
lengths, but not by the ―drop-in‖ effect. The observed rotations for sections prior to drop-in erection were small
enough to allow for this erection approach should adequate control of the previously erected sections be provided.
The initial erection sequencing study findings were validated via additional investigations involving final curved and
skewed designs. Vertical and radial deflections were the indexes used to evaluate the adequacy of the erection
approaches that were applied to the selected final designs. Normalized deflections, consisting of maximum
deflections from a given construction scenarios normalized with respect to maximums from the recommended
preliminary study erection scenario (paired inner), were studied. Findings indicated that:
Curved
o Girder vertical deflections were decreased when paired-girder erection methods were
used.
o The paired inner erection was preferred for structures with severe curvature.
Skewed
o Erection methods examined herein did not show appreciable influence on skewed bridge
behavior.
Studies of the effects of web out-of-plumbness on construction performance examined curved and skewed structures
with a 1% imposed out-of-plumbness, the maximum permissible value for plate girders according to the Bridge
Welding Code (AASHTO/AWS, 2004). The out-of-plumbness was introduced to the girder webs in a fashion that,
178
for the curved bridges, increased any anticipated twist in the girders and, for the skewed bridges, increased any
anticipated global twisting during deck placement. Girder webs were intentionally tilted 1% of their depth along
their entire length, with cross-frame geometries being modified accordingly. Static, sequential analysis was
performed to examine the out-of-plumbness effects at each stage, with the paired inner erection method again being
adopted since it was the recommended approach from the sequencing studies. Vertical and radial deflections were
primarily compared, and at each stage the maximum girder vertical deflections and maximum radial deflections at
splice locations were examined for the out-of-plumb cases, with those maximum deflections being normalized by
the maximum deflections for the plumb web case. To examine out-of-plumbness effects on stresses, comparisons of
maximum Von Mises stresses in the cross frames were also examined. Findings indicated that:
Curved
2004).
o Exceeding the 1% limit of the web out-of-plumbness can result in slightly higher vertical and
lateral deformations and also stresses. However, the effects of horizontal curvature on these
parameters are much larger than those from the web out-of-plumbness.
Skewed
2004).
o Exceeding the 1% limit of the web out-of-plumbness can result in slightly higher vertical
deformations and stresses. However, the effects on lateral deformations are more pronounced. As
a result, the effects of web-out-of plumbness, which could be beneficial to ensure web-plumb at
the completion of construction, should be considered during the erection of structures having a
skew less than 70 .
Temporary construction shoring placement was examined for curved and skewed final structures by applying the
paired inner erection method to structures that were shored at: splice locations only; splice locations as well as at
0.4L of the span for exterior spans and and/or 0.5L for interior spans; and at 0.75L of the first span and at splice
locations for the second span of the two-span bridges. In addition to studying the effects of these shoring placement
schemes, the effects of support settlement in conjunction with the use of shoring was studied by examining critical
curved and skewed structures in conjunction with three levels of support settlement, 0.001L, 0.002L, and 0.004L,
where L is the span length between the abutment and pier, and having these settlement amounts occurring at either
one shoring support or two support locations. Girder deformations and cross-frame Von Mises stresses were
normalized against the web-plumb case and used to examine the shoring and settlement effects. Findings indicated
that:
Curved
o While not shoring the superstructure during erection is certainly preferred from a costeffectiveness standpoint, should shoring be deemed necessary during construction in addition to
shoring at splice locations, placing one temporary support at 0.4L could significantly reduce girder
deflections, leading to a more constructible condition.
o When adding additional shoring at 0.4L was not feasible, placing one support at 0.75L between
the abutment and splice location could reduce girder vertical deflections by more than 75% when
compared to the deflections of placing one support at the splice location.
o Attempts should be made to limit support settlement to less than 0.001L (L=span length) during
o Small differences were observed between one support settling and two supports settling.
Skewed
o While not shoring the superstructure during erection is certainly preferred from a costeffectiveness standpoint, should shoring be deemed necessary during construction, in addition to
shoring at splice locations, placing one temporary support at 0.4L could significantly reduce girder
deflections, leading to a more constructible condition.
179
o
o
o
Attempts should be made to limit support settlement to less than 0.001L (L=span length) during
Two supports settling had a larger impact on stresses than one support settling.
The effects of inconsistently detailing cross-frame members during construction, studied earlier by Chavel and Earls
(2006), were re-examined for final curved and skewed bridge designs by numerically studying paired-inner erection
methods with and without inconsistently detailed cross frames. Temperature changes were placed into cross-frame
members to either shorten or lengthen those members to the anticipated geometric conditions that would result when
the girders were detailed to be web-plumb under one construction condition (e.g. no load) and cross frames were
mistakenly detailed to be web-plumb under another condition (e.g. full dead load). Initial models were created in
SAP2000 to establish what the inconsistently detailed cross-frame lengths would be, then sequencing models were
run in ABAQUS to study their effects. Again, normalized girder deformations and cross-frame Von Mises stresses
were used as the primary methods to assess the detailing effects. Findings indicated that:
Curved
o Inconsistent cross-frame detailing increased vertical and radial deflections.
o Cross-frame stresses were appreciably increased as a result of the locked-in stresses due to the
o Irrespective of the radius and cross-frame spacing, inconsistent cross-frame detailing should be
avoided for all curved structures to mitigate locked-in stresses and increased deformations. The
spacings.
Skewed
o Inconsistent cross-frame detailing did not have a crucial impact on girder vertical deflections.
o Inconsistent cross-frame detailing increased lateral deflections.
o Cross-frame stresses were appreciably increased as a result of the locked-in stresses due to
o Irrespective of the skew and cross-frame spacing, inconsistent cross-frame detailing should be
avoided for all skewed structures to mitigate locked-in stresses and increased deformations. The
effects of inconsistent detailing can more pronounced in bridges with larger cross frame spacings.
The effects of replacing cross frames with diaphragms on construction behavior was examined via the study of three
replacement scenarios: (1) diaphragms replacing cross frames at the abutments only, (2) diaphragms replacing cross
frames at the abutments and piers, and (3) diaphragms replacing cross frames at the abutments, piers, and at midspan. Paired-girder erection was used to study their effects, and the girder deformations and cross-frame Von Mises
stresses, which were normalized against similar values for the case where no diaphragms were present, were again
used to assess the effects on construction response. Findings indicated:
Curved
o Replacing cross frames with solid plate diaphragms did not severely affect bridge vertical and
radial deflections and cross-frame stresses during construction, irrespective of the radius and
cross-frame spacing.
curved bridges that were studied did not adversely affect or appreciably benefit their construction
behavior. Nevertheless, using diaphragms at the mid-span locations can cause slightly higher
deformations for the girders when compared to the use of cross frames at those locations.
Skewed
o Placing solid plate diaphragms in skewed bridges slightly increased deformations due to the self
weight of the diaphragms, but did not severely affect cross-frame stresses.
skewed bridges did not adversely affect or appreciably benefit their construction behavior.
180
However, using diaphragms at the mid-span locations can cause slightly higher deformations for
the girders when compared to the use of cross frames at those locations.
Temperature effects on construction response were studied by examining their influence on girder deflections and
cross-frame stresses when a 50 °F temperature increase occurred during deck placement. Paired-girder erection was
used, and values were normalized with respect to no temperature change occurring as the deck was placed. Findings
indicated that:
Curved
and stresses for all of the radii and cross-frame spacings studied.
Skewed
and stresses for all of the skew angles and cross-frame spacings studied.
These findings are summarized in Table 34.
Table 34. Summary Table.
DM 4/ 408/
Proposed Language
Report
BD
DM4 Section
2, Index
DM4 2.5.3.1P
DM4
C2.5.3.1P
Section
Single Inner Erection - Erecting a single girder from inner
radius to outer radius.
Single Outer Erection - Erecting a single girder from outer
radius to inner radius.
Paired Inner Erection - Erecting two girders from inner
radius to outer radius.
Paired Outer Erection - Erecting two girders from outer
radius to inner radius.
Should the contractor deem that temporary falsework is
necessary for the construction of curved and skewed steel
bridges, the following guidelines should be used for its
placement:
When temporary falsework is needed for a span, it
shall be placed at locations to reduce splice rotations
and girder vertical deflections..
The stability of the structure supported by temporary
falsework shall be evaluated.
While using no temporary falsework is desirable from a
cost-effectiveness perspective, should the designer and/or
contractor deem that falsework is needed to ensure that a
curved or skewed steel bridge is constructible, it should
initially be placed near splice locations. When girder
vertical deflections are still a concern, an additional
temporary support should be placed as close as possible to
the location of maximum vertical deflection of the span
(approximately 0.4 L from an abutment, where L is the
span length, for side spans and 0.5 L for intermediate
spans) to reduce girder deflections. For the side spans,
when adding multiple temporary supports is not feasible,
placing one support near 0.75L from the abutments is
suggested.
181
5.4
5.4
5.4
DM4 2.5.3.2P
DM4
C2.5.3.2P
The following guidelines should be used for girder
erection of horizontally curved steel I-girder structures:
Should adequate crane capacity be available,
paired girder erection approaches are preferred.
When the radius of the curved structure is less
than 300 feet, it is recommended that girders be
placed from inner radius to outer radius.
An analysis shall be performed to ensure that the
structure is stable for all stages of construction
and that supports necessary to maintain stability
have been provided.
The following guidelines should be used for girder
erection of skewed steel I-girder structures:
structure is stable for all stages of construction
and that supports necessary to maintain stability
have been provided.
Paired girder erection, as opposed to single erection,
requires fewer temporary supports for the erected
segments during all stages of construction. However for
bridges with an odd number of girders, at least one girder
line must be erected by itself.
5.2
5.2
Girder erection from inner radius to outer radius, when
compared to the opposite direction, can result in slightly
smaller deformations for the girders for all stages of
construction which, in turn, means the structure is more
constructible. This effect is more pronounced in severely
curved structures (i.e., radius less than 300 ft).
Stability of partial and completed girders at various stages
of erection is the responsibility of the contractor, as
specified in Publication 408 Section 1050.3(c).
DM4 Section
For construction of straight skewed bridges, paired
erection does require a smaller number of temporary
supports but offers no other substantial benefits over a
single erection approach with respect to deformations.
PennDOT Publication 408, Section 1050, ―Steel Bridge
Superstructure,‖ 2007.
2, Reference
DM4 Section
Span Length - The distance between supports along the
centerline of the girder web.
3, Index
DM4 3.12.8P
When falsework is used, an analysis should be performed
to check its settlement effects on response during
construction. As a minimum, the following scenarios
should be considered for the analysis:
Settlement of single and multiple temporary supports .
182
5.4
A minimum settlement of one thousandth of the span
length should be used.
DM4
C4.6.1.2.1
DM4
C4.6.2.2.1
DM4
The selected refined method of analysis for a structure
curved in plan must provide an accurate prediction of
behavior, both during construction and while in-service.
While the method of analysis that is selected is at the
discretion of the designer, a superstructure modeling
technique that represents the girder webs and concrete
deck using shell elements and other major superstructure
components using beam elements provides an acceptable
compromise between reduced computation times provided
by grillage analogy models and increased accuracy
provided by more sophisticated three-dimensional finite
element models.
This does not mandate a 3-D analysis, but does mean a
special analysis of the cross-frame must be provided in
order to account for the differential deflections which
occur across a cross-frame. This analysis should
accurately account for cross frame member geometry and
stiffness. Should a grillage analogy model be used,
accurate representation of cross frame stiffness should be
established via special analysis of representative frames.
Should a more sophisticated 3-D analysis be used, models
can be constructed following the technique recommended
in C4.6.1.2.1 for structures curved in plan.
Diaphragm members in horizontally curved and skewed
bridges may be used at support locations.
5.6
6.7.4.1
DM4
C6.7.4.1
DM4
C6.7.4.1
DM4
C6.7.4.1
bridges may be used at support locations. When support
lines are skewed less than 70 degrees, intermediate
diaphragms or cross-frames shall be placed normal to the
girders in contiguous or discontinuous lines. When cross
frames and diaphragms are normal to web near skewed
supports, adequate girder restraint shall be provided.
Placement of cross frames parallel to the skew has been
shown to induce significant localized lateral bending near
support locations (AASHTO Subcmte on Bridges &
Structures, 2010).
5.6
Solid plate diaphragms can be used at support locations in
horizontally curved and skewed bridges. Replacing cross
frames with solid plate diaphragms at abutment and pier
locations has been shown to not adversely affect or
appreciably benefit deformations during construction. For
other intermediate locations along the bridge spans,
diaphragms may cause higher stresses and deformations in
the bridge structures during construction when compared
to the use of cross frames.
5.6
183
DM4 6.7.4.2
DM4
C6.7.4.2
DM4
Cross frames and steel girders in curved and skewed
bridges should both be detailed so that the webs are plumb
under a specified loading condition.
In curved and skewed bridges, cross-frame stresses can
increase appreciably as a result of locked-in stresses
caused by inconsistent detailing. This increase may be
more pronounced in bridges with larger cross frame
spacings and for cross frames near the bridge supports.
For bridges with skew angles between 90 and 70 ,
develop shop drawings which detail all webs plumb when
girders are erected and diaphragms connected. For curved
bridges and skewed bridges with skew angles less than
70 , develop shop drawings and erection procedures
which detail all webs plumb after the full dead load (self
weight of all structural and non-structural components,
not including weight of the future wearing surface) is
applied. See Article 6.7.2 and BC-754, ―Steel
Diaphragms.‖
(f) Temperature changes as prescribed in 3.12.2.1.1.
5.5
The effects from temperature change on the curved and
skewed bridges are mostly functions of the girder support
conditions. When minimum required restraint necessary
for girder(s) global stability (i.e., prevention of global
buckling of the girder or collection of girders) is
provided, the applied temperature change may not have
an appreciable impact on overall bridge deflections and
stresses.
American Association of State Highway and
Transportation Officials (AASHTO), Subcommittee on
Bridges and Structures, Agenda Item 14,
http://bridges.transportation.org/Pages/2010AgendaIte
ms-SacramentoCA.aspx, 2010.
5.7
5.5
5.7
6.10.3.2.5.1P
DM4
C6.10.3.2.5.1P
DM4 Section
6, Reference
BD-620M
PennDOT Bridge Construction Standards, BC-754, Steel
Diaphragms. 2006.
ADDITIONAL LATERAL STABILITY CRITERIA FOR
SKEWED STEEL BRIDGES
6. CROSS FRAME INCONSISTENT DETAILING
SHOULD BE AVOIDED. FOR BRIDGES WITH
SKEW ANGLES BETWEEN 90 AND 70 ,
DEVELOP SHOP DRAWINGS WHICH DETAIL
ALL WEBS PLUMB WHEN GIRDERS ARE
ERECTED AND DIAPHRAGMS CONNECTED.
FOR SKEWED BRIDGES WITH SKEW ANGLES
LESS THAN 70 , DEVELOP SHOP DRAWINGS
AND ERECTION PROCEDURES WHICH DETAIL
ALL WEBS PLUMB AFTER THE FULL DEAD
LOAD (SELF WEIGHT OF ALL STRUCTURAL
AND NON-STRUCTURAL COMPONENTS, NOT
INCLUDING WEIGHT OF THE FUTURE
WEARING SURFACE) IS APPLIED.
184
5.5
ADDITIONAL LATERAL STABILITY CRITERIA FOR
CURVED STEEL BRIDGES
408 Section
1050
6. CROSS FRAME INCONSISTENT DETAILING
SHOULD BE AVOIDED. FOR ALL CURVED
BRIDGES DEVELOP SHOP DRAWINGS AND
ERECTION PROCEDURES WHICH DETAIL ALL
WEBS PLUMB AFTER THE FULL DEAD LOAD
(SELF WEIGHT OF ALL STRUCTURAL AND
NON-STRUCTURAL COMPONENTS, NOT
INCLUDING WEIGHT OF THE FUTURE
WEARING SURFACE) IS APPLIED.
(c) Erection. References to Section of Division II,
AASHTO Standard Specifications for Highway
Bridges (date as indicated), and PennDOT Design
Manual – Part 4 are identified by the abbreviation
AASHTO or PennDOT DM4, followed by the
section number, e.g., AASHTO 11.6.4.
5.2 and
5.4
2. Falsework Design and Construction. AASHTO
11.2.2, 11.6.1, Section 105.03(c) and PennDOT
DM4 2.5.3.1P.
3. Erection Procedure. AASHTO 11.6.4 and Penn DOT
DM4 6.10.3.2.5.1P.
A recommended implementation plan for the findings from this study and subsequent suggested findings in
Appendix D would consist of:
PennDOT reviews the findings obtained from the project summarized herein and suggests corresponding
modification to relevant PennDOT documents.
PennDOT selects suggested revisions that the Department views as acceptable for inclusion in the
PennDOT documents.
PennDOT initiates the process to include those items in future publications.
PennDOT reviews the findings from other relevant projects that will be completed after publication of this
report for inclusion in future PennDOT documents. These projects include NCHRP Project 12-79.
Suggested modifications from the present work that PennDOT views as acceptable can be integrated with
any findings from the other relevant projects.
Development and implementation by PennDOT of training tools to assist with implementation of any
changes in relevant PennDOT documents.
185
7
REFERENCES
American Association of State Highway and Transportation Officials. (2007). LRFD Design Specifications,
Washington, D.C.
American Association of State Highway and Transportation Officials. (2003). Guide Specifications for Horizontally
Curved Steel Girder Highway Bridges, Washington, D.C.
American Association of State Highway and Transportation Officials/American Welding Society, (2004). Structural
Welding Code-Steel, 19th Edition, Miami, FL.
Chavel, B.W. and Earls, C.J. (2006), ―Construction of a Horizontally Curved Steel I-Girder Bridge. PartII:
Inconsistent Detailing‖, ASCE Journal of Bridge Engineering, Vol.11, (1), pp. 91–98.
Hiltunen, D.R., Johnson, P.A., Laman, J.A., Linzell, D.G., Miller, A.C., Niezgoda, S.L., Scanlon, A., Schokker, A.J.
and Tikalsky, P.J. (2004). Interstate 99 Research, Contract No. SPC 020S78, Pennsylvania Department of
Transportation, October, 324 pp.
Galambos, T.V., Hajjar, J.F., Leon, R.T., Huang, W-H, Pulver, B.E., and Rudie, B.J. (1996). ―Stresses in Steel
Curved Girder Bridges,‖ Minnesota Department of Transportation Report No. MN/RC – 96/28, August.
Linzell, D.G., Laman, J.A., Bell, B., Bennett, A., Colon, J., Lobo, J., Norton, E. and Sabuwala, T. (2003). Prediction
of Movement and Stresses in Curved and Skewed Bridges, University-Based Research, Education and Technology
Transfer Program; Agreement No. 359704, Work Order 79. Final Report, Pennsylvania Department of
Transportation, March, 192 pp.
Linzell, D.G., Nadakuditi, V.P. and Nevling, D.L. (2006). Prediction of Movement and Stresses in Curved and
Skewed Bridges, PennDOT/MAUTC Partnership, Work Order No. 2, Research Agreement No. 510401, Final
Report, Pennsylvania Department of Transportation, September, 86 pp.
Linzell, D.G., Seo, J. and Coughlin, A. (2008). Guidelines for Analyzing Curved and Skewed Bridges and
Designing them for Construction: Updated Literature Search Report, The Thomas D. Larson Pennsylvania
Transportation Institute Report No. PTI 2008-15, August, 31 pp.
Linzell, D.G., Nevling, D.L., and Seo, J. (2008). Guidelines for Analyzing Curved and Skewed Bridges and
Designing them for Construction: Numerical Modeling Report, The Thomas D. Larson Pennsylvania Transportation
Institute Report No. PTI 2008-16, August, 57 pp.
Mintab,(2007). Users Manuals; Version 15.1.20.0. Minitab, Inc., State College, Pa.
National Weather Service, National Oceanic and Atmospheric Administration. <http://www.nws.noaa.gov>
Pennsylvania Department of Transportation (2007). PennDOT Design Manual – Part 4, Structures (Publication
15M).
186
8
APPENDIX A
Updated Literature Search Report
187
PENNDOT RESEARCH
GUIDELINES FOR ANALYZING CURVED AND SKEWED BRIDGES
AND DESIGNING THEM FOR CONSTRUCTION
LITERATURE SEARCH REPORT
Work Order No. PSU009
Intergovernmental Agreement, No. 510602
August 15, 2008
By D. G. Linzell, J. Seo and A. Coughlin
PENNSTATE
188
Transportation Research Building
(814) 865-1891 www.pti.psu.edu
GUIDELINES FOR ANALYZING CURVED AND SKEWED BRIDGES AND DESIGNING
THEM FOR CONSTRUCTION
LITERATURE SEARCH REPORT
Work Order No. PSU-009
Intergovernmental Agreement No. 510602
Prepared for
Commonwealth of Pennsylvania
Department of Transportation
By
Daniel G. Linzell, Ph.D., P.E.
Junwon Seo
Andrew Coughlin
December 15, 2008
PTI 2008-15
This work was sponsored by the Pennsylvania Department of Transportation and the U.S.
Department of Transportation, Federal Highway Administration. The contents of this report
reflect the views of the authors, who are responsible for the facts and the accuracy of the data
presented herein. The contents do not necessarily reflect the official views or policies of either
the Federal Highway Administration, U.S. Department of Transportation, or the Commonwealth
of Pennsylvania at the time of publication. This report does not constitute a standard,
specification, or regulation.
TABLE OF CONTENTS
8.1 INTRODUCTION ................................................................................................................171
8.2 GENERAL DESIGN GUIDELINE AND LITERATURE SEARCH..................................171
8.3 PROPOSED DESIGN FOR CONSTRUCTION DOCUMENT TEMPLATE ....................176
REFERENCES ............................................................................................................................177
190
8.1
INTRODUCTION
The modern transportation industry encounters an increasing use of curved and skewed I-girder/beam bridges for a
number of reasons. These types of bridges are becoming more common as highway infrastructure is increasingly
rebuilt atop existing structures to handle increasing traffic volumes or new interchange geometries within the context
of urban settings. In particular, curved I-girder/beam bridges have the ability to change direction within each span
and thus are ideal structures for applications such as highway interchanges or to connect existing roadways where
abutments cannot be relocated for physical or economic reasons. Additionally, specification of the curved structure,
while generating more superstructure costs in terms of materials and engineering, actually reduces the structure’s
cost through the elimination of interior supports, significant deck overhangs, and expensive right-of-way
acquisitions. Similarly, skewed I-girder/beam bridges are also useful when roadway alignment changes are not
feasible or economical because of the topography of the site and also at particular areas where environmental impact
is an issue.
However, curved and skewed I-girder/beam bridges tend to significantly deflect and rotate out of plane under the
action of gravity. In particular, these bridge types of construction prove to be a challenge where each girder tends to
rotate under its own weight and any additional load (e.g., concrete deck) applied perpendicular to the plane of
curvature. Oftentimes, the construction of such bridges is more complex, and more detailed consideration of the
construction process is required when compared to constructing corresponding straight steel I-girder bridges. As
mentioned above, this is because unlike a straight steel I-girder, curved and skewed girder bridge construction must
control not only the vertical displacement of the I-girders but also the out-of-plane displacements such that structural
components, including cross frames, can be erected with limited difficulty. Therefore, varied construction strategies
attempt to actively counteract the girders’ tendency to deflect and rotate out of plumb.
To achieve better understanding of the effects of design, fabrication, and construction on the geometry and load
distribution in a curved or skewed I girder/beam bridge system, further study on such bridges is required. Therefore,
this project is intended to continue examination of the behavior of curved and skewed I-girder bridges during
construction with the intention of developing criteria to assist the process. The specified objectives of this project are
presented as follows:
(1) Continue to develop and maintain acquisition capabilities for instruments on two structures in the I-99
corridor – #207, a horizontally curved, steel, I-girder bridge, and #314, a skewed, pre-stressed, concrete
bridge.
(2) Develop, examine, and reduce data produced from these structures as needed.
(3) Continue examination of numerical model accuracy for curved and skewed steel I-girder bridges and
select appropriate model types and software for continued application.
(4) Extend numerical studies to examine prevalent issues affecting curved and skewed steel I-girder bridge
behavior during construction.
(5) Develop relevant guidelines for curved and skewed steel I-girder bridges during construction.
8.2
GENERAL DESIGN GUIDELINE AND LITERATURE SEARCH
This report serves as an addendum to literature searches completed in association with earlier projects funded by
PennDOT. Resulting information can be found in reports published in association with those projects (Linzell et al.
2003; Hiltunen et al. 2004; Linzell et al. 2006) and, for horizontally curved steel bridges, in other publications
(Zureick et al. 1994; Hall et al. 1999; Linzell et al. 2004; Kulicki et al. 2006). The focus of the literature search
completed herein was to expand previous findings and to expand the body of the literature search to include specific
items identified as additional focus areas by the sponsor during development of the current project scope.
191
Recent work completed in association with the construction response of curved steel bridges has been published in
association with the Curved Steel Bridge Research Project (CSBRP) by the AASHTO/NSBA Steel Bridge
Collaboration and by others. The recent research work has largely focused on revision of existing specifications
(AASHTO 2005), development of model types and prototype software packages that can be used to predict
construction response (Chang 2006; Chang and White 2006), publication of qualitative recommendations and
representative plan sets to assist with the erection of steel structures (AASHTO/NSBA 2003, 2006), examination of
cross frame detailing and erection decisions on final geometry at the completion of superstructure erection (Chavel
and Earls 2006a, 2006b), study of lifting requirements for curved girder segments coupled with the development of
strength and slenderness limits for non-composite curved I-girder flanges that account for web contributions
(Madhavan 2006), examining web out-of-plumbness and resulting problems related to deck pour sequencing, cross
bracing forces (Howell and Earls 2007), and examination of the effect of bearing type on curved steel bridge
behavior (Samaan et al. 2002).
Recent research related to skewed bridge response during construction has been more limited and was largely
summarized in past submittals (Linzell et al. 2003); however, revisions to existing specifications have also occurred
(AASHTO 2005). Work by the AASHTO/NSBA Steel Bridge Collaboration above has also attempted to address
skewed steel bridge construction issues in largely qualitative form, and other research has been published related to
steel superstructure response during deck placement (Choo et al. 2005; Norton et al. 2003). A few studies related to
bearing types and their influence on skewed steel bridge behavior have also been published (Samaan et al. 2002).
More detailed information related to design guidelines research for curved and skewed I-girder/beam bridges is
incorporated into Section 2.1, while more detailed information related to additional research in the additional focus
areas is incorporated into Section 2.2.
8.2.1
General Design Provisions and Guidelines
For bridge designers in Pennsylvania, PennDOT language related to curved and skewed steel I-girder bridges is
largely found in Sections 2, 3, 4, 6, and 14 of Part B of PennDOT Publication 15M, Design Manual Part 4
(PennDOT 2007). These sections refer to and/or supplement similar sections in the AASHTO LRFD Bridge Design
Specifications (AASHTO 2004), with Section 2 providing serviceability information, Section 3 general design
information, Section 4 analysis information, Section 6 strength design information, and Section 14 bearing detailing
information. As a result of the approach and the utilization of references to AASHTO, specific information related to
construction is limited and qualitative in nature with specific mention of construction effects limited to consideration
of uplift forces that could occur.
PennDOT Publication 408/2007 (PennDOT 2007) contains construction specifications for PennDOT projects with
specific references to steel superstructures occurring in Section 1050. This section largely deals with bearing
placement and tolerances with no specific differentiation between straight bridges and those of irregular geometry.
Qualitative mention of erection procedures is also made, again with no differentiation between straight bridges and
curved and skewed structures.
Research related to curved steel I-girder bridges has historically focused on in-service behavior and development of
design criteria to address this response. Curved Steel Bridge Research Project (CSBRP) research in this area
resulted in extensive changes in Section 6 of the 2005 Interim AASHTO LRFD Bridge Design Specifications
(AASHTO 2005) which were incorporated into PennDOT Publication 15M, Design Manual Part 4 (PennDOT
2007). Until recently, work related to construction response of these structures was largely completed in association
with I-99 corridor construction or via other research completed by the PI (Hiltunen et al. 2006; Linzell et al. 2002,
2004; Nevling et al. 2006; Zureick et al. 2000; Linzell 1999). However, since the submission of earlier I-99 corridor
reports, there has been valuable research by others that examines construction response.
Ongoing work by the AASHTO/NSBA Steel Bridge Collaboration has resulted in the publication of a number of
documents related to steel bridge construction. Of these documents, the Guidelines for Design for Constructability,
the Steel Bridge Erection Guide Specification, and corresponding sample erection plans (AASHTO/NSBA 2003,
2006) contain the most specific references and guidance for curved steel I-girder bridge construction. While these
192
documents contain a large amount of valuable information relevant to the scope of this project, they are also largely
qualitative in nature.
Previously discussed revisions to the AASHTO LRFD Bridge Design Specifications do include more information
related to construction issues for curved steel I-girder bridges resulting from CSBRP work (AASHTO 2005), with
designers being directed to pay close attention to deflection and stability issues during construction.
Research related to skewed steel bridge construction response has been much more limited, and historical work was
summarized in previous submittals to PennDOT (Linzell et al. 2003). While the 2005 AASHTO Interims do discuss
preferential cross frame placement schemes for steel superstructures as a function of skew angle and the need to
address differential deflections during design, the most comprehensive recent publications attempting to address
construction response are the aforementioned AASHTO/NSBA Steel Bridge Collaboration documents. In these
documents, discussion of skewed construction aspects is also largely qualitative in nature and focuses on the effects
of skew on superstructure deformations and how to counteract those effects and placement and orientation of cross
frames in skewed superstructures.
8.2.2
Updated Literature Search
Topics that have been updated through the current literature search includes research focusing on the following areas
as they relate to curved and skewed bridge construction behavior: (1) development of model types and prototype
software packages; (2) web-plumbness; (3) bearing type and restraint; (4) cross frames behavior; (5) temporary
shoring placement and settlement; and (6) global temperature change. Summaries of each topic are provided below.
8.2.2.1
Development of model types and prototype software packages
A number of modeling strategies for design and construction analysis of curved I-girder bridge systems ranging
from modified line-girder analyses to finite element approaches have been recently published. Chang et al. (2006)
conducted research on a representative full-scale composite curved I-girder bridge, tested at the FHWA TurnerFairbank Highway Research Center (TFHRC) and utilized the bridge for assessment of different modeling
approaches. The predictions of key lateral and vertical displacements, reactions, cross frame forces, and major-axis
and flange lateral bending stresses were evaluated with respect to the test bridge. They discussed the efficacy of the
various approaches. In addition, this research focused on the development and validation of a prototype software
package that focuses on predicting construction response of curved steel I-girder bridge systems during girder
erection and the deck pour. Coupled with the creation of this tool was the development of expanded 2D modeling
approaches for curved I-girder bridges during construction. Topkaya and Williamson (2003) presented a variety of
computational efforts related to predicting the behavior of horizontally curved trapezoidal steel girders during
erection and before the concrete deck hardened to form a closed section. The program they developed, UTrap, was
formulated based on the finite element method to account for construction sequencing and was designed to be
computationally efficient and easy for bridge designers to use. It was reported that the developed software was able
to accurately capture girder stresses during construction (Topkaya et al. 2003).
8.2.2.2
Effects of web-plumbness
In horizontally curved steel girder bridges, the girders have a tendency to deform out-of-plumb as they are put under
vertical load because of load eccentricity with respect to the curved girder supports locations. In horizontally curved
bridges with smaller radii of curvature and longer spans, web out-of-plumbness is a concern because it can increase
stresses in girder flanges, cause fabrication problems with cross bracing and deck pouring, increase cross bracing
forces, and create the perception of safety problems (Howell and Earls 2007). Flanges of curved girders have been
shown to have stress increases as high as 23% due to vertical and lateral bending caused by out-of-plumbness.
Vertical bending stresses can also increase in parts of the flanges because of added eccentricity from the neutral axis
caused by the radii of curvature. Lateral bending results from a weak axis component of the vertical loads on an
out-of-plumb girder. These effects should be included in any stress analysis of girders (AASHTO 2003).
193
Differential deflections and out-of-plumbness can cause problems with cross frame detailing; inconsistent detailing
and geometric inconsistencies can result (Domalik et al. 2005), and increased cross frame forces can occur (Howell
and Earls 2007). To alleviate this problem, one of several techniques can be used. The most common method used
when web plumbness problems arise during fabrication is to impose outside forces by means of cranes and/or
jacking to set the member in a position where the cross frame can be placed. This, however, can be costly, can
impose stresses that the superstructure was not designed to carry, and at times may require forces above the capacity
of the equipment used for construction (Chavel and Earls 2006b). As an alternative, the engineer of record can
specify vertical and twist camber to result in a near plumb condition at a specified stage of construction (e.g., no
load, steel self weight, full dead load) agreed upon by the steel fabricator (Chavel and Earls 2006b,
AASHTO/NSBA 2003). In addition, some deflection problems have been shown to be mitigated by using a suitable
erection sequence (Bell and Linzell 2007).
Safety perception problems can likely be mitigated by keeping the out-of-plumbness to a minimum when the bridge
is in operation. Since its effect is very small after the deck hardens, care should be taken in the design stage to
ensure that the girders will be near plumb after the deck is poured. This care should be taken especially in cases of
long span and small radius of curvature combinations (Howell and Earls 2007).
While limited studies have attempted to examine the effects of web-plumbness on curved bridge construction
behavior, there have been no reported studies on the effects of web-plumbness on skewed steel bridge behavior
during construction.
8.2.2.3
Effects of bearing type and restraint
The type and arrangement of bearings for a bridge superstructure are important considerations in bridge design. For
a curved continuous girder bridge, the support conditions for the bridge superstructure may significantly influence
the distribution of maximum stresses, reactions, and shear forces, as well as the bridge’s natural frequencies and
mode shapes. However, current design practices in the U.S. recommend very few guidelines for bearing
arrangements and types. A few papers related to bearing types for curved and skewed steel bridges have been
published (Samaan et al. 2002; Tindal and Yoo, 2003).
Samaan et al. (2002) described an extensive study carried out using an experimentally calibrated finite element
model in which curved, continuous prototype bridges were analyzed to determine their structural response. Six
different types and arrangements of support bearings were studied to determine their effects on the maximum stress
and reaction distributions, as well as on the natural frequencies of such bridges. The results were used to suggest the
most favorable bearing arrangement and type.
Tindal and Yoo (2003) examined bearing effects on skewed steel highway bridges. Three bearing orientation cases,
representative of current AASHTO LRFD design practices, were considered, and parametric studies were
conducted. Hypothetical bridges were designed for a range of different span lengths, section depths, widths, and
skews. Each bridge model was tested under all three bearing orientation case parameters, and the relative influence
of each parameter on behavior was discussed.
8.2.2.4
Effects of cross frames
Because of the coupling action of the vertical bending and torsion, curved and skewed girders are subject to
significant rotations. During construction, the noncomposite steel girder must support the wet concrete and steel
weight in addition to other construction loads, such as the weight of screed, formwork, and other items. There have
been large relative deflections observed between girders on curved and skewed girder bridges under steel selfweight
and during deck placement that make it difficult to maintain the specified final cambers and superelevations and to
form and key in the construction joints. Since it is not practical to increase the girder stiffness simply to minimize
the relative deflections and rotations, either cross frames, lateral bracing, temporary shoring, or a combination of
these items is considered. Because the cross frames are not only expensive but can adversely affect the fatigue
behavior of the steel girder, the number of cross frames should be minimized (Norton et al. 2003).
194
However, there are currently few design guidelines available for intermediate cross frames in curved steel girder
bridges. There have been a few publications related to cross frames’ effects on curved bridges, with much of the
focus being curved box girders (Memberg, 2002; Kim and Yoo, 2006). Memberg (2002) presented a procedure for
the design of intermediate external diaphragms that recommended designing these members to carry forces that were
more than ten times higher than those measured experimentally. Kim and Yoo (2006) evaluated the effects of cross
frames on deck unevenness during construction by modeling hypothetical twin-box girder bridges using ABAQUS.
A deck unevenness ratio was defined that quantified the degree of unintended deck slope caused by relative
superstructure deflections and rotations. Additional relevant research associated with examinations of the influence
of cross frame detailing and specific erection decisions on final superstructure geometry for a curved I-girder bridge
has also been published (Chavel and Earls 2006a, 2006b). Findings from this work were used to emphasize the need
for designers to explicitly specify the dead load state used to detail the curved superstructure to prevent fit-up
problems during erection.
In addition, there have been a few papers related to cross frames’ effects on skewed steel I-girder bridges
(Azizinamini et al. 1995; Wang et al. 2008). Azizinamini et al. (1995) conducted analytical and experimental
investigations on the influence that cross frames having different configurations had on load resisting capacity.
Results indicated that for steel bridges with small skew, the presence of cross frames had little influence on behavior
after construction. Cross frames were shown not only to be unnecessary after construction but were stated to be, to a
degree, harmful as they tried to prevent the tendency of the girders to separate and consequently transferred
restraining forces to the girder webs, which could cause fatigue cracking. Wang et al. (2008) outlined cross frame
placement requirements for bridges with skewed supports when considering primarily stability; however, stiffness
and strength were also addressed. Two orientations of the intermediate cross frame were considered, and as a result
of this research, stability cross frame requirements for steel girders were presented as a function of support geometry
and bracing orientation.
8.2.2.5
Temporary shoring placement and settlement
Many horizontally curved steel I-girder bridges utilize shoring during construction. It is often assumed that the
bridge superstructure (girders and cross bracing) will be erected in a specified no-load condition. To achieve this,
temporary shoring (or falsework) is implemented to control horizontal and vertical deflections during erection
(Chavel and Earls 2006). Other erection procedures involve erecting the girders in pairs, with cross bracing being
erected on the ground and then the pair being lifted into place. This can result in fewer horizontal and vertical
alignment problems than single girder erection techniques, but shoring has been shown to further reduce alignment
problems (Bell and Linzell 2007). When project constraints allow shoring to be utilized, it will likely provide a
benefit by eliminating fit-up problems that can arise during construction. Hydraulic jacks mounted on the shoring
can further correct any alignment problems that may arise during fabrication and deck pouring. Chavel and Earls
(2006) have suggested that some misalignments during construction thought to be from girder deflections are in fact
due to fabrication and detailing errors since finite element models were unable to replicate the misaligned response.
Publications related to shoring location or support settlement effects on curved and skewed structure construction
response could not be identified.
8.2.2.6
Global temperature change during deck placement
Very little information exists about the effects that global temperature changes during deck placement have on steel
bridge behavior be they straight, curved, or skewed bridges. NCHRP’s Steel Bridge Erection Practices publication
(2005) identifies horizontal and vertical deflections caused by temperature changes as a problem that several
contractors have reported, but no solutions are given.
The research that exists has studied a skewed steel bridge for the purpose of validating finite element models. In the
process of monitoring deflections and stresses induced into an actual bridge during placement of the deck, a finite
element model showed reduced accuracy as the placement progressed throughout a single day. When thermal
effects were added, model predictions improved significantly (Choo et al. 2005).
195
8.2.3
Summary
The latest general PennDOT and federal design provisions, along with updated research and work in additional
specified areas related to curved and skewed bridge systems during construction, were summarized. Updates to
general design provisions have occurred but have been largely qualitative in nature. Updates to already established
curved and skewed bridge research areas related to construction have largely consisted of further examination and
development of tools to accurately predict construction response. Additional topics that were included in the present
literature search consisted of web out-of-plumbness, bearing type and restraint, cross frames; temporary shoring
placement and settlement, as well as global temperature change and those topics as they related to curved and
skewed bridge construction, had somewhat limited information available in the literature.
8.3
PROPOSED DESIGN FOR CONSTRUCTION DOCUMENT TEMPLATE
As discussed and included in the current project scope document, the format in which findings from this work would
be presented would permit incorporation into relevant PennDOT specification documents should PennDOT feel that
modifications suggested by Penn State personnel are warranted. As was discussed in the scope document,
modifications will focus on: (1) Sections 2, 3, 4, and 6 within PennDOT Publication 15M, Design Manual Part 4
(PennDOT 2002); (2) Section 1050.3 in PennDOT Specifications Publication 408/2007 (PennDOT 2007); and
PennDOT Standard Drawing Set BD-620M (PennDOT 2004). A proposed initial draft containing possible
modifications to each of the aforementioned documents was included in the scope and is reproduced in the pages
that follow. Additional suggestions for modifications as a result of research work that has occurred since initiation
of the project have also been incorporated into the following pages. Any suggested modifications or changes to the
initial draft document since initiation of the project are shown in 12-Point font, in bold and italics. It should
be emphasized that all suggested modifications provided herein are preliminary and that extensive changes could
occur as the project progresses.
196
8.4
REFERENCES
AASHTO (2004). AASHTO LRFD Bridge Design Specifications. American Association of State Highway and
Transportation Officials, Washington, D.C.
AASHTO (2005). 2005 Interim AASHTO LRFD Bridge Design Specifications. Customary U.S. Units. American
Association of State Highway and Transportation Officials, Washington, D.C.
AASHTO/NSBA Steel Bridge Collaboration (2003, 2006). Guidelines for Design for Constructability G12.1.
American Association of State Highway and Transportation Officials, Washington, D.C.
Azizinamini, A., Kathol, S., and Beacham, M. W. (1995). ―Influence of Cross Frames on Load Resisting Capacity of
Steel Girder Bridges,‖ Engineering Journal, 32(3), 107-116.
Bell, B. J., and Linzell, D. G. (2007). ―Erection Procedure Effects on Deformations and Stresses in a Large Radius,
Horizontally Curved, I-Girder Bridge.‖ J. Bridge Eng., 12(4), 467-476.
Chang, C. J., White, D. W., Beshah, F. and Wright, W. (2006). ―Design Analysis of Curved I-Girder Bridge Systems
- An Assessment of Modeling Strategies,‖ Annual Proceedings, Structural Stability Research Council, 349-369.
Chang, C. J. (2006). ―Construction Simulation of Curved Steel I-Girder Bridges,‖ Thesis in Civil Engineering,
Georgia Institute of Technology, Atlanta, GA, USA.
Chavel, B. W., and Earls, C. J. (2006a). ―Construction of a Horizontally Curved Steel I-Girder Bridge. Part I:
Erection Sequence.‖ J. Bridge Eng., 11(1), 81-90.
Chavel, B. W., and Earls, C. J. (2006b). ―Construction of a Horizontally Curved Steel I-Girder Bridge. Part II:
Inconsistent Detailing.‖ J. Bridge Eng., 11(1), 91-98.
Choo, T, Linzell, D. G., Lee, J. I., and Swanson, J. A. (2005). ―Response of a continuous, skewed, steel bridge
during deck placement.‖ J. of Constructional Steel Research, 61, 567-586.
Dey, G. (2001). ―Bridging the curve: Design and fabrication issues affecting economy and constructability,‖
Bridgeline, 11(1). HDR Engineering, Omaha, NE.
Domalik, D. E., Shura, J. F., and Linzell, D. G. (2005). ―Design and Field Monitoring of Horizontally Curved Steel
Plate Girder Bridge.‖ Transportation Research Record 1928, Transportation Research Board, Washington, D.C.,
83–91.
Hall, D. H., Grubb M. A., and Yoo, C. H. (1999). Improved Design Specifications for Horizontally Curved Steel
Girder Highway Bridges, National Cooperative Highway Research Program, Research Report 424, Transportation
Research Board, Washington, D.C.
Hiltunen, D. R., Johnson, P. A., Laman, J. A., Linzell, D. G., Miller, A. C., Niezgoda, S. L., Scanlon, A., Schokker,
A. J., and Tikalsky, P. J. (2004). Interstate 99 Research, Final Report, Pennsylvania Transportation Institute, Report
No. PTI 2005-02, October, 324 pp.
Howell, T. D. (2006). ―On the influence of web out of plumbness on horizontally curved steel I-girder bridge
serviceability during construction,‖ Thesis in Civil Engineering, Univ. of Pittsburgh, Pittsburgh, PA.
Howell, T. D., and Earls, C. J. (2007). ―Curved Steel I-Girder Bridge Response during Construction Loading:
Effects of Web Plumbness,‖ Journal of Bridge Engineering, 12(4), 485-493.
197
Kim, K. S., and Yoo, C. H. (2006). ―Effects of external bracing on horizontally curved box girder bridges during
construction,‖ Engineering Structure, 28 1650–1657.
Kulicki, J. M., Wassef, W. G., Kleinhans, D. D., Yoo, C. H., Nowak, A. S. and Grubb, M. (2006). Development of
LRFD Specifications for Horizontally Curved Steel Girder Bridges, National Cooperative Highway Research
Program, Research Report 563, Transportation Research Board, Washington, D.C.
Linzell, D. G. (1999). ―Studies of a Full-Scale Horizontally Curved Steel I-Girder Bridge System under SelfWeight,‖ Thesis in Civil Engineering, Georgia Institute of Technology, Atlanta, GA, USA.
Linzell, D. G., Laman, J. A., Bell, B., Bennett, A., Colon, J., Lobo, J., Norton, E., and Sabuwala, T. (2003).
Prediction of Movement and Stresses in Curved and Skewed Bridges, University-Based Research, Education and
Technology Transfer Program; Agreement No. 359704, Work Order 79. Final Report, Pennsylvania Department of
Transportation, March, 192 pp.
Linzell, D. G., Hall, D. H., and White, D. W. (2004). ―A Historical Perspective on Horizontally Curved I-Girder
Bridge Design in the United States,‖ ASCE Journal of Bridge Engineering, 9(3), 218-229.
Linzell, D. G., Leon, R. T., and Zureick, A. H. (2004). ―Experimental and Analytical Studies of a Horizontally
Curved Steel I-Girder Bridge during Erection,‖ ASCE Journal of Bridge Engineering, 9(6), 521-530.
Linzell, D. G., Nadakuditi, V. P. and Nevling, D. L. (2006). Prediction of Movement and Stresses in Curved and
Skewed Bridges, Final Report, Pennsylvania Transportation Institute, Report No. PTI 2007-05, November, 93 pp.
Madhavan, M., and Davidson, J. S. (2004). ―Elastic buckling of centerline-stiffened plates subjected to a linearly
varying stress distribution.‖ Proceedings of the Twenty-second Southeastern Conference on Theoretical and Applied
Mechanics, August 15-17, Tuskegee Alabama.
Memberg, M. A. (2002). ―A design procedure for intermediate external diaphragms on curved steel trapezoidal box
bridges,‖ MS thesis. Austin (TX): University of Texas.
NCHRP. (2005). Steel Bridge Erection Practices, National Cooperative Highway Research Board Synthesis 345,
Transportation Research Board, Washington, D.C.
Nevling, D., Linzell, D., and Laman, J. (2006). ―Examination of Level of Analysis Accuracy for Curved I-Girder
Bridges through Comparisons to Field Data,‖ ASCE Journal of Bridge Engineering, 11(2), 160-168.
Norton, E. K., Linzell, D. G., and Laman, J. A. (2003). ―Examination of the Response of a Skewed Steel Bridge
Superstructure During Deck Placement,‖ Transportation Research Record 1845, 66-75.
Pennsylvania Department of Transportation (2007), Publication 408/2007.
Pennsylvania Department of Transportation (2004), BD-620M, Standard, Steel Girder Bridges Lateral Bracing
Criteria and Details.
Pennsylvania Department of Transportation (2004), Publication 15M, Design Manual Part 4.
Samaan, M., Sennah, K., and Kennedy, J. B. (2002). ―Positioning of bearings for curved continuous spread-box
girder bridges,‖ Canadian Journal of Civil Engineering, 29: 641–652.
Tindal, T. T., and Yoo, C. H. (2003). ―Thermal Effects on Skewed Steel Highway Bridges and Bearing Orientation,‖
J. Bridge Eng., 8(2), 57-65.
Topkaya, C., and Williamson, E. B. (2003). ―Development of computational software for analysis of curved girders
under construction loads,‖ Computers and Structures 81, 2087-2098.
198
Wang, L., and Helwig, T. A. (2008). ―Stability Bracing Requirements for Steel Bridge Girders with Skewed
Supports,‖ J. Bridge Eng., 13(2), 149-157.
Zureick, A., Naqib, R., and Yadlosky, J. M. (1994). Curved Steel Bridge Research Project, Interim Report I:
Synthesis, HDR Engineering, Inc., Publication Number FHWA-RD-93-129, December.
Zureick, A., Linzell, D., Leon, R. T., and Burrell, J. (2000). ―Curved Steel Bridges: Experimental and Analytical
Studies,‖ Engineering Structures, 22(2).
199
9
APPENDIX B
Numerical Modeling Report
200
PENNDOT RESEARCH
GUIDELINES FOR ANALYZING CURVED AND SKEWED BRIDGES
AND DESIGNING THEM FOR CONSTRUCTION
FINAL INTERIM REPORT: NUMERICAL MODELING
Work Order No. PSU009
Intergovernmental Agreement, No. 510602
September 9, 2009
By D. G. Linzell, D. L. Nevling and J. Seo
PENNSTATE
201
(814) 865-1891 www.pti.psu.edu
GUIDELINES FOR ANALYZING CURVED AND SKEWED BRIDGES AND DESIGNING
THEM FOR CONSTRUCTION
FINAL INTERIM REPORT: NUMERICAL MODELING
Work Order No. PSU-009
Intergovernmental Agreement No. 510602
Prepared for
Commonwealth of Pennsylvania
Department of Transportation
By
Daniel G. Linzell, Ph.D., P.E.
Deanna L. Nevling
Junwon Seo
September 9, 2009
PTI 2008-16
This work was sponsored by the Pennsylvania Department of Transportation and the U.S.
Department of Transportation, Federal Highway Administration. The contents of this report
reflect the views of the authors, who are responsible for the facts and the accuracy of the data
presented herein. The contents do not necessarily reflect the official views or policies of either
the Federal Highway Administration, U.S. Department of Transportation, or the Commonwealth
of Pennsylvania at the time of publication. This report does not constitute a standard,
specification, or regulation.
202
TABLE OF CONTENTS
9.1 INTRODUCTION ................................................................................................................184
9.2 STRUCTURE DESCRIPTIONS ..........................................................................................184
9.3 MODELING TECHNIQUES ...............................................................................................196
9.4 COMPARISONS AND MODIFICATIONS ........................................................................199
9.5 ADDITIONAL EVALUATIONS ........................................................................................222
9.6 CONCULSIONS...................................................................................................................233
9.7 REFERENCES .....................................................................................................................234
203
9.1
Introduction
This report summarizes modeling development and verification completed in association with Work Order
009, Guidelines for Analyzing Curved and Skewed Bridges and Designing Them for Construction. The
work order is a continuation of earlier projects and information related to modeling efforts completed in
association with those projects can be found in earlier reports (Linzell et al. 2003; Hiltunen et al. 2004;
Linzell et al. 2006).
The intent of the modeling portion of the project is to further develop and select models to assist with
parametric studies being completed in association with future project tasks. Findings from Task 1
(Literature Search) of the current project coupled with detailed analyses and accuracy evaluations for one
curved bridge, Structure #207, were used to complete any additional major modifications required to
improve model accuracy. The effectiveness of these modifications was further evaluated by comparing
model predictions to information obtained during the construction of other curved and skewed bridges. The
curved bridges include Structure #7A, a structure that was examined and summarized in past submittals to
PennDOT (Linzell et al. 2003; Hiltunen et al. 2004; Linzell et al. 2006) and the I-79 Missing Ramps
Bridge. The skewed structure is Structure #28, another bridge that was examined and summarized in past
submittals to PennDOT (Linzell et al. 2003).
The sections that follow provide information on the structures that were modeled and discuss the modeling
techniques examined with the current effort and with other efforts completed in association with this study.
In addition, they summarize comparisons, modifications, and additional evaluations that were completed.
9.2
Structure Descriptions
As was discussed above, a total of four structures were utilized for model validation and modification.
Structure descriptions and erection and instrumentation information are provided herein, and, if applicable,
references are provided to previous PennDOT submittals containing additional information.
9.2.1
Structure #207
Structure #207 is the bridge that was used for the majority of the model validation and modification steps
performed in association with this project. Additional discussion of the bridge design and erection is
provided in past submittals to PennDOT (Linzell et al. 2003; Hiltunen et al. 2004; Linzell et al. 2006). A
brief design, erection, and instrumentation summary is provided herein.
Structure #207 is located in Centre County, PA, at the intersection of Interstate 99 and United States Route
322 East. The two-span continuous bridge contains five singly-symmetric steel plate girders. It spans a
total distance of 146.53 m (480.75 ft) with Span 1 measuring 65.38 m (214.50 ft) and Span 2 ranging 81.50
m (266.25 ft). The bridge centerline radius of curvature is 585 m (1921 ft), and the girders are spaced at
3.25 m (10.67 ft) center to center. Four field splices are located along the bridge.
All girder webs are 2.74 m (108.00 in) deep. Top and bottom flanges dimensions vary along the girder
length. More detail on the girder dimensions at select locations and representative structural drawings can
be found in previous PennDOT submittals (Linzell et al. 2006).
Structure #207 was constructed using a single girder erection procedure, working from interior girder (G5)
toward exterior girder (G1). Span 1 was erected along with a portion of Span 2 (to Field Splice 3) prior to
any girder placement within Span 2. The contractor utilized temporary shoring towers under G4 and G5
and tie-downs at the abutments and piers. Three cranes were used to place the girders, and a boom truck
was used to place the cross-frames. Deadmen were affixed to G2 at the completion of erection to provide
additional radial restraint prior to the deck pour. Bolts were installed throughout the process but were not
tightened until all girders and cross-frames were in place. Superstructure erection took 10 days to complete.
Initially, the contractor attempted to tighten bolted girder field splices by lifting a single girder splice at a
time, which resulted in undesired movement of adjacent, tightened field splices. Therefore, all five girders
204
were lifted at a given field splice simultaneously using the temporary shoring towers, and this problem was
mitigated. As tightening and aligning progressed, minor alignment problems (top of girder elevations) were
observed, and two splices needed to be reassembled.
Placement of the 229-mm (9-ft), reinforced concrete deck was conducted in four stages. Deck
reinforcement consisted of two epoxied rebar mats. Main reinforcing steel was oriented radially and
consisted of #5 bars spaced 152 mm (6 in) on center. Transverse reinforcement consisted of #5 bars spaced
at 279 mm (11″) in positive moment regions, and #5 and #6 bars alternating at 127 mm (5″) in the negative
moment region. To minimize deck cracking in the negative moment regions, concrete in the positive
moment region of Span 1 was placed first, followed by the positive moment region of Span 2. The negative
moment region over the pier was poured during the third stage, with the blockouts at Abutments 1 and 2
poured during the final stage. A pump truck was used to deliver concrete onto the bridge, and a finishing
machine and work bridges were used to place and finish concrete to the desired elevation and cross-slope.
Each consecutive placement step was performed within two days of the previous step. Parapets were added
after the entire deck had achieved 28-day strength. Additional detail on the erection process can be found in
previous submittals to PennDOT (Linzell et al. 2003; Linzell et al. 2006).
Penn State University personnel instrumented the structure using a total of 65 vibrating wire strain gauges
and 4 vibrating wire tiltmeters, and 10 laser targets were used to monitor the bridge during construction.
The majority of the instruments were placed on the exterior girders and on cross-frames identified as
critical elements in the curved system from preliminary analyses. Additional details on the preliminary
analyses and the instrumentation can be found in previous submittals to PennDOT (Linzell et al. 2003;
Hiltunen et al. 2004; Linzell et al. 2006).
9.2.2
Structure #7A
Structure #7A was used for some earlier model validation studies and, as such, is described in previous
submittals to PennDOT (Linzell et al. 2003). A brief design, erection, and instrumentation summary is
provided herein.
Structure #7A is one of two side-by-side horizontally curved composite steel I-girder bridges constructed at
an I-99 interchange over Park Avenue in Centre County, Pennsylvania. It is a six-span structure whose
cross-section consists of five singly symmetric plate girders spaced at 2.97 m (9.75 ft) with radii varying
from 585.3 m to 597.2 m (1920 ft to 1959 ft). The girders are composed of stiffened web plates with a
constant depth and thickness and flange plates of varying dimensions. Girders are braced radially using
cross-frames made up of WT sections, and no lateral bracing system was included in the original
construction plans.
The complete structure consists of 2 three-span continuous units spanning a total distance of 530.1 m (1739
ft) along the roadway’s construction line. The eastern of these, which is the focus of this work, is
composed of spans designated Spans 4, 5, and 6. More detail on the structure dimensions can be found in
previous PennDOT submittals (Linzell et al. 2003).
A construction plan was prepared by the original contractor that called for erection of girders in pairs, with
the single girder line having the highest radius of curvature being placed after the other four girder lines
(two pairs) had been erected in a given span. Each girder pair was preassembled on the ground with crossframes fully bolted and tightened, and the units were then raised incrementally from splice to splice.
Erection was initiated with placement of Span 4 steel, beginning at Pier 4 and working toward Pier 3. Prior
to the erection phase, falsework was mounted to Pier 4 to support hydraulic jacks that stabilized and
adjusted the system during construction to achieve the design ―no-load‖ geometry prior to deck placement.
Originally, construction was to be accomplished using four cranes, and no temporary shoring towers would
be employed.
Halfway through erection of Span 4, twisting of G2 and G3 caused undesired deformations and made it
impossible to continue without revising construction measures and/or adding bracing. The proposed
205
solution included implementing a single support tower in Span 4, which was placed approximately 36 m
(118.11 ft) east of Pier 3 and would allow cranes to be released to resume erection. Complications
continued to occur during the erection procedure, and upon completion of Span 4 and a cantilevered portion
of Span 5, the superstructure was surveyed and shown to be severely out of lateral alignment from the
intended ―no-load‖ geometry, with horizontal misalignments of 274 mm (10.8 in) at Field Splice #11 in
Span 4 and 351 mm (13.8 in) at Field Splice #14.
A second contractor was hired to realign the superstructure and complete the erection of the remaining
spans. Once the design geometry of the erected portion of the bridge was attained, upper lateral bracing
was inserted between the fascia and first interior girders, and steel placement continued using a revised plan
that called for the erection of single girder lines rather than girder pairs, placing the girder line with the
largest radius of curvature first and working inward. Furthermore, the revised scheme required the
implementation of temporary support towers in all spans. More detail on the intended and final erection
plans can be found in previous PennDOT submittals (Linzell et al. 2003).
A limited field monitoring program was instituted during construction of Structure 7A. Data were collected
during two phases of steel erection: (1) realignment of the previously erected portion in Span 4 and a
portion of Span 5 and (2) completion of steel erection in Spans 5 and 6. Field survey data tracking
deformations at various stages throughout both processes were collected by a third party. More detail on the
instrumentation and data collection plans can be found in previous PennDOT submittals (Linzell et al.
2003).
9.2.3
Missing Ramps Bridge
The Missing Ramps Bridge is a seven-span continuous composite steel plate girder bridge with a total
length of 452.93 m (1486 ft). The superstructure utilizes five ASTM A709 Grade 50 singly symmetric steel
plate girders that are 2.29 m (90 in) deep and spaced at 2.13 m (7 ft) center to center. Flange plate
dimensions vary along the length of the girders. Spans 1, 2, 3, 4, 5, 6, and 7 measure 55.93 m (183 ft 6 in),
72.39 m (237 ft 6 in), 72.39 m (237 ft 6 in), 72.39 m (237 ft 6 in), 55.50 m (182 ft 1in), 63.27 m (207 ft 7
in), 68.88 m (226 ft), and 63.7 m (209 ft) along the arc, respectively, and the radius of curvature to the
center girder is 253.14 m (830 ft 6 in). Lateral bracing is provided using both X- and K-shaped crossframes containing WT sections oriented radially along both spans. For shipping purposes, each girder
consists of five sections that are bolted at 11 field splices. Table 35 provides some information on girder
geometry. Figure 232 and Figure 233 detail plan views and typical superstructure cross-sections.
206
Table 35. Girder Lengths, Radii, and Field Splice Locations (see Figure 1).
Span
(1)
Girder
ID
(2)
G1
G2
1
and
2
G3
G4
G5
G1
G2
3
and
4
G3
G4
G5
G1
G2
5
and
6
G3
G4
G5
G1
G2
7
G3
G4
G5
Radius
L1
L2
L3
L4
L5
m (ft)
m (ft)
m (ft)
m (ft)
m (ft)
m (ft)
First
Span
m (ft)
Second
Span
m (ft)
Total
m (ft)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
248.87
37.04
39.24
30.25
19.20
N/A
54.69
70.79
125.48
(816.50)
(121.52)
(128.75)
(99.24)
(63.00)
N/A
(179.43)
(232.24)
(411.67)
251.00
37.51
39.24
30.85
19.20
N/A
55.16
71.39
126.55
(823.50)
(123.06)
(128.75)
(101.23)
(63.00)
N/A
(180.97)
(234.23)
(415.20)
253.14
37.98
39.24
31.46
19.20
N/A
55.63
71.98
127.61
(830.50)
(124.59)
(128.75)
(103.22)
(63.00)
N/A
(182.51)
(236.17)
(418.68)
255.27
38.44
39.24
32.07
19.20
N/A
56.10
72.61
128.70
(837.50)
(126.13)
(128.75)
(105.21)
(63.00)
N/A
(184.05)
(238.21)
(422.26)
257.40
38.91
39.24
32.67
19.20
N/A
56.57
73.21
129.78
(844.50)
(127.67)
(128.75)
(107.20)
(63.00)
N/A
(185.59)
(240.20)
(425.79)
248.87
19.20
30.25
39.01
36.84
N/A
70.79
54.26
125.05
(816.50)
(63.00)
(99.24)
(128.00)
(120.85)
N/A
(232.24)
(178.02)
(410.26)
251.00
19.20
(30.85)
39.01
37.30
N/A
71.39
54.73
126.12
(823.50)
(63.00)
101.23
(128.00)
(122.39)
N/A
(234.23)
(179.56)
(413.79)
253.14
19.20
31.46
39.01
37.77
N/A
71.98
55.20
127.18
(830.50)
(63.00)
(103.22)
(128.00)
(123.93)
N/A
(236.17)
(181.09)
(417.26)
255.27
19.20
32.07
39.01
38.24
N/A
72.61
55.67
128.27
(837.50)
(63.00)
(105.21)
(128.00)
(125.46)
N/A
(238.21)
(182.63)
(420.84)
257.40
19.20
32.67
39.01
38.71
N/A
73.21
56.14
129.35
(844.50)
(63.00)
(107.20)
(128.00)
(127.01)
N/A
(240.20)
(184.17)
(424.37)
248.87
14.63
27.11
34.75
38.10
14.63
61.86
67.36
129.22
(816.50)
(48.00)
(88.95)
(114.00)
(124.99)
(48.00)
(202.95)
(220.99)
(423.94)
251.00
14.63
27.65
34.75
38.68
14.63
62.39
67.94
130.33
(823.50)
(48.00)
(90.70)
(114.00)
(126.89)
(48.00)
(204.70)
(222.89)
(427.59)
253.14
14.63
28.18
34.75
39.25
14.63
62.93
68.51
131.44
(830.50)
(48.00)
(92.46)
(114.00)
(128.78)
(48.00)
(206.46)
(224.78)
(431.24)
255.27
14.63
28.72
34.75
39.83
14.63
63.46
69.09
132.55
(837.50)
(48.00)
(94.21)
(114.00)
(130.68)
(48.00)
(208.21)
(226.68)
(434.89)
257.40
14.63
29.25
35.00
40.41
14.63
64.00
69.67
133.66
(844.50)
(48.00)
(95.96)
(114.83)
(132.57)
(48.00)
(209.96)
(228.57)
(438.53)
248.87
24.38
38.90
N/A
N/A
N/A
63.29
N/A
63.29
(816.50)
(80.00)
(127.63)
N/A
N/A
N/A
(207.63)
N/A
(207.63)
251.00
24.38
39.15
N/A
N/A
N/A
63.54
N/A
63.54
(823.50)
(80.00)
(128.46)
N/A
N/A
N/A
(208.46)
N/A
(208.46)
253.14
24.38
39.41
N/A
N/A
N/A
63.79
N/A
63.79
(830.50)
(80.00)
(129.30)
N/A
N/A
N/A
(209.30)
N/A
(209.30)
255.27
24.38
39.66
N/A
N/A
N/A
64.05
N/A
64.05
(837.50)
(80.00)
(130.13)
N/A
N/A
N/A
(210.13)
N/A
(210.13)
257.40
24.38
39.92
N/A
N/A
N/A
64.30
N/A
64.30
(844.50)
(80.00)
(130.96)
N/A
N/A
N/A
(210.96)
N/A
(210.96)
207
N
FIELD SPLICE 3
FIELD SPLICE 2
ABUTMENT #1
5.59m (18' 4 3/16'')
TYP
' 9 1/2'')
3m (19
PIER #2
TYP
6.0
FIELD SPLICE 1
PIER #1
G1
G2
G3
G4
G5
2.13m (7'-0") TYP
L4
L3
L2
SPAN 2
L1
SPAN 1
(a) Spans 1 and 2
N
PIER #2
FIELD SPLICE 6
FIELD SPLICE 5
8'
5.55m (1
2 1/2'')
TYP
PIER #4
2.13m (7'-0") TYP
PIER #3
6.03m (19' 9 1/2'') TYP
FIELD SPLICE 4
G1
G2
G3
G4
G5
L4
L3
L1
SPAN
L2
4
SPAN 3
(b) Spans 3 and 4
N
PIER #4
FIELD SPLICE 10
PIER #6
FIELD SPLICE 9
FIELD SPLICE 8
'') TYP
2.13m (7'-0") TYP
8' 10
5.74m (1
FIELD SPLICE 7
PIER #5
'') TYP
5.75m (18' 10 7/16
G1
G2
G3
G4
G5
L5
L4
L3
L1
SPAN 6
L2
SPAN 5
(c) Spans 5 and 6
208
N
PIER #6
FIELD SPLICE 11
ABUTMENT #2
5.79m (19' 0'') TYP
G1
G2
G3
G4
G5
2.13m (7'-0") TYP
L2
L1
SPAN 7
(d) Span 7
Figure 232. Plan Views, Missing Ramps Bridge.
1'-8 1/4" (0.51m)
32'-4 1/2" (9.87m) OUT TO OUT PARAPETS
30'-8 1/2" (9.36m) CURB TO CURB
15'-0" (4.57m) LANE
4'-0" (1.22m)
SHOULDER
13" (33.0cm) CONCRETE DECK
10'-0" SHOULDER
(3.05m)
VARIES%
VARIES%
G1
G3
G2
G4
1'-8 1/4"
(0.51m)
PARAPET
G5
4 SPACES @ 7'-0" (2.13m) = 28'-0" (8.53m)
2'-2 1/4" (0.66m)
2'-2 1/4" (0.66m)
(a) End Section
1'-8 1/4" (0.51m)
4'-0" (1.22m)
SHOULDER
32'-4 1/2" (9.87m) OUT TO OUT PARAPETS
30'-8 1/2" (9.36m) CURB TO CURB
15'-0" (4.57m) LANE
13" (33.0cm) CONCRETE DECK
10'-0" SHOULDER
(3.05m)
VARIES%
VARIES%
G2
G1
G3
G4
1'-8 1/4"
(0.51m)
PARAPET
G5
4 SPACES @ 7'-0" (2.13m) = 28'-0" (8.53m)
2'-2 1/4" (0.66m)
2'-2 1/4" (0.66m)
(b) Interior Section
Figure 233. Typical Cross-Sections, Missing Ramps Bridge.
209
The Missing Ramps Bridge was constructed using a single girder erection procedure, working from either
the exterior girder (G5) toward interior girder (G1) or from the interior girder toward the exterior girder,
depending upon the location. A combination of cranes and temporary bents was used to lift and hold girder
segments during the process. Due to the existence of an expansion joint over Pier 4, the girder erection
sequence was separated into Units 1 and 2 as shown in Figure 234 and Figure 235, which re-create the
planned sequence. The actual erection sequence initiated with Unit 2 and ended with Unit 1 with some
modifications occurring during the initial stages of the Unit 2 erection. These modifications are detailed
below.
210
PIER
#1
FIELD
SPLICE
1
ABUTMENT #1
A
TEMP BENT 1
(a) Stage 1
PIER
#1
FIELD
SPLICE
1
ABUTMENT #1
A
FIELD
SPLICE
2
B
TEMP BENT 1
(b) Stage 2
FIELD
SPLICE
6
PIER
#3
PIER
#4
G
TEMP BENT 3
(c) Stage 3
FIELD
SPLICE
5
FIELD
SPLICE
6
PIER
#3
PIER
#4
F
G
TEMP BENT 3
(d) Stage 4
FIELD
SPLICE
1
ABUTMENT #1
PIER
#1
A
PIER
#2
FIELD
SPLICE
2
FIELD
SPLICE
4
FIELD
SPLICE
5
E
B
TEMP BENT 1
TEMP BENT 2
(e) Stage 5
FIELD
SPLICE
1
ABUTMENT #1
PIER
#1
A
FIELD
SPLICE
2
PIER
#2
FIELD
SPLICE
3
B
FIELD
SPLICE
4
D
FIELD
SPLICE
5
E
TEMP BENT 1
TEMP BENT 2
(e) Stage 6
FIELD
SPLICE
1
ABUTMENT #1
A
PIER
#1
PIER
#2
C
B
FIELD
SPLICE
4
D
FIELD
SPLICE
2
E
FIELD
SPLICE
3
TEMP BENT 1
TEMP BENT 2
(e) Stage 7
Figure 234. Unit 1, Girder Erection Details, Missing Ramps Bridge.
211
FIELD
SPLICE
5
FIELD
SPLICE
7
PIER #4
FIELD
SPLICE
8
H
G
H
TEMP BENT 4
(a) Stage 1
FIELD
SPLICE
7
PIER #4
H
G
FIELD
SPLICE
9
FIELD
SPLICE
8
H
I
TEMP BENT 4
(b) Stage 2
FIELD
PIER
SPLICE
7
PIER #4
H
G
FIELD
SPLICE
7
#4
H
FIELD
SPLICE
TEMP BENT 4
7
H
H
G
H
FIELD
SPLICE
9
8
I
H
G
PIER #4
FIELD
FIELD
SPLICE
9
SPLICE
FIELD
SPLICE
8
FIELD
SPLICE
10
I
J
FIELD
SPLICE
9
FIELD
SPLICE
8
FIELD
SPLICE
10
I
J
TEMP BENT 4
TEMP BENT 5
TEMP BENT 4
PIER #4
PIER #4
FIELD
SPLICE
7
FIELD
SPLICE
7
H
(c) Stage 3
FIELD
SPLICE
8
FIELD
SPLICE
8
H
G
G
TEMP BENT 5
H
PIER #6
FIELD
SPLICE
10
I
H
J
TEMP BENT 4
FIELD
SPLICE
FIELD
10
SPLICE
11
J
I
TEMP BENT 4
FIELD
SPLICE
9
FIELD
SPLICE
9
K
TEMP BENT 5
TEMP BENT 5
(d) Stage 4
L
PIER #4
G
H
FIELD
SPLICE
7
PIER #5
FIELD
SPLICE
8
H
I
TEMP BENT 4
FIELD
SPLICE
9
FIELD
SPLICE
10
J
PIER #6
FIELD
SPLICE
11
ABUTMENT
#2
K
TEMP BENT 5
(e) Stage 5
Figure 235. Unit 2, Girder Erection Details, Missing Ramps Bridge.
The girder erection schedule for Unit 1 involved 7 stages as shown in Figure 234. For Stage 1, Temporary
Bent 1 was erected, and all bearings on Abutment 1 were configured. All bearings were blocked to prevent
translation and movement in the longitudinal direction. Segment A was erected using an exterior girder
(G5) toward interior girder (G1) procedure with cross-frames being fit between adjacent girders after they
were erected. For Stage 2, all bearings on Pier 1 were configured, and an interior girder (G1) to exterior
girder (G5) procedure was used, again with cross-frames fit between adjacent girders after their placement.
For Stage 3, Temporary Bent 3 was erected, and bearings on Pier 4 were configured and blocked. Interior
to exterior girder erection was used. For Stage 4, all bearings on Pier #3 were configured and an interior to
exterior girder sequence was used. For Stage 5, Temporary Bent 2 was erected, and interior to exterior
212
PIER #6
K
FIELD
SPLIC
11
girder erection was used. For Stage 6, bearings at Pier 2 were blocked, and interior to exterior erection was
used. For Stage 7, the final Unit 1 stage, interior to exterior erection was used.
Unit 2 had five planned stages as shown in Figure 235. For Stage 1, the original planned procedure shown
in the figure, which involved using a combination of cranes and Temporary Bent 4 to lift, position, and
stabilize the girders, was replaced with a procedure that eliminated Bent 4 (circled in figure) and utilized
cranes only. In addition, Field Splice 7 (circled in figure) was eliminated from those segments. All other
erection tasks from the original plan, including configuring and blocking the Pier 4 bearings, erecting
segments from exterior toward interior girder and fitting cross-frames between girders after they were in
place, were followed. At the completion of Stage 1, it was decided to return to the original erection plan
and utilize a combination of temporary shoring towers and cranes. Therefore, Stage 2 involved
configuration and blocking of bearings at Pier #5 and interior to exterior girder erection as shown in Figure
235. For Stage 3, Temporary Bent 5 was built as originally planned, and girder segments were erected from
interior toward exterior girder (G5). For Stage 4, bearings at Pier 6 were configured and blocked to prevent
movement, and interior toward exterior girder erection was used. The final stage began with configuration
and blocking of the Abutment 2 bearings and utilized interior to exterior girder erection.
In addition to cranes used to place the girders, a boom truck was used to install the cross-frames. Bolts were
installed throughout the girder erection process but were not tightened until all girders and cross-frames
were in place.
The deck placement sequence was also divided between Units 1 and 2. As shown in Figure 236, for Unit 1,
deck in the positive moment regions (1 and 2) was placed first. To minimize deck cracking in the negative
moment regions, region 3 over the piers was poured during the final stage. The approach is similar for Unit
2 shown in Figure 237, where deck in positive moment regions 4 and 5 was placed first followed by
negative moment region 6 over the piers. A pump truck was used to deliver concrete to the bridge, and a
finishing machine and work bridges were used to place and finish concrete to the desired elevation and
cross-slope.
During girder erection, PennDOT personnel tracked girder web rotations at each cross-frame connection
point. A number of measurements were taken, and each involved evaluating the amount of radial
translation of the top of the girder web toward or away from the center of the curvature during girder
erection. This information was used by Penn State personnel to assist with computer model validation as
reported in the following sections.
213
ER
#4
12
5'
6''
(3
8.2
5m
)
2
PI
#3
57
'0
'' (
17
.3
7m
)
58
'0
3
'' (
17
.6
8m
)
ER
PI
1
PI
ER
#2
)
7m
.9
35
'(
0'
'
8
11
3
'6
62
ER
#1
PI
8'
11
1
AB
M
UT
'6
62
2
T#
EN
'' (
'' (
)
5m
.0
19
m)
.05
19
)
7m
5.9
(3
0''
3
'0
58
1
'0
58
'' (3
5' 6
12
'' (
'' (
m)
.68
17
m)
.68
17
)
5m
8.2
Figure 236. Unit 1 Deck Placement Sequence, Missing Ramps Bridge.
214
TME
2
NT #
R #6
15
.2
4m
)
15
9'
PIE
0''
(4
8.
46
m
)
4
ABU
50
'0
'' (
6
R #5
'
94
R #4
'6
49
4
(
9''
)
8m
.8
28
)
9m
.8
19
'' (
3
''
65
6
PIE
)
2m
.1
20
5
PIE
'' (
'0
66
'' (
m)
.09
15
m)
.62
(48
6''
'
9
15
Figure 237. Unit 2 Deck Placement Sequence, Missing Ramps Bridge.
9.2.4
Structure #28
In similar fashion to Structure #7A, Structure #28 was used for some earlier model validation studies and,
as such, is described in previous submittals to PennDOT (Linzell et al. 2003). A brief design, erection, and
instrumentation summary is provided herein.
Structure #28 is a single-span composite steel-plate girder bridge located at an interchange between I-99
and Pennsylvania State Route 150. The bridge is 74.45 m (244 ft 3 in) in length with a skew of
approximately 55 degrees. A total of seven girders are used to support the structure, with each girder being
constructed using 17.5 mm by 2400 mm (11/16 in by 94 in) web plates and flange plates of varying
dimensions. Girders are braced using ―X‖ shaped cross-frames between the supports and ―K‖ shaped
cross-frames at the supports. Cross-frames are staggered near each abutment. The fascia girders (G1 and
G7) and their adjacent interior girders (G2 and G6) are supported by non-guided expansion bearings at the
abutments. Remaining girders utilize guided expansion bearings at the west abutment and fixed bearings at
the east abutment. Additional details on structure dimensions can be found in previous PennDOT
submittals (Linzell et al. 2003).
215
Girders were erected with web plates out of plumb at the abutments and at midspan. Girders were
fabricated in the plumb position, and the cross-frames were used to force the webs out of plumb by an
amount equal and opposite to the anticipated rotation after placement of the deck.
Concrete deck placement began at the east abutment and proceeded perpendicular to the centerline of the
bridge using two screeds that were staggered to attempt to place wet concrete parallel to the skew. Screed
rails were attached to G1, G4, and G7. More details on the erection and deck placement schemes can be
found in previous PennDOT submittals (Linzell et al. 2003).
Structure #28 was monitored by Penn State personnel during the deck pour, which occurred overnight.
Longitudinal strains and girder displacements were measured using strain transducers and linear variable
differential transformers (LVDTs). Strain transducers attempted to measure strains in girder flanges and in
individual cross-frame members at select locations and the LVDTs measuring lateral displacements of the
girder webs at the abutments. Global geometric data were also collected from traditional surveys
performed before and after the deck placement process and using a three-dimensional laser scanner system.
Additional details on the instrumentation can be found in previous submittals to PennDOT (Linzell et al.
2003).
9.3
Modeling Techniques
This section details modeling techniques selected for the study. Included in this discussion are summaries
of examined model construction approaches, including discretization levels, constitutive models, and
support conditions, along with a discussion of analysis techniques. If applicable, references are provided to
previous PennDOT submittals containing relevant information.
9.3.1
Model Construction
The types of models examined included variations of grillage (two-dimensional) finite element analysis
approaches and variations of three-dimensional approaches that included a mix of line, shell, and solid
elements. For certain model types, detailed discussion of their characteristics are provided in previous
submittals to PennDOT, and, therefore, only brief summaries are provided herein.
9.3.1.1
Grillage (2D)
Grillage models used for predicting construction response were developed in SAP2000 in association with
earlier model validation studies and, as such, are largely described in previous submittals to PennDOT
(Linzell et al. 2006). Therefore, a brief summary is provided herein.
For these models, the steel superstructure was modeled as a true grillage; however, in an attempt to more
accurately represent deck load distribution onto the steel superstructure, when the concrete deck was being
placed, it was modeled using shell elements. As a result, the final model could not be called a true grillage
as classically defined but was referred to as a ―modified‖ grillage model. Composite action was simulated
for stages that required it using rigid links. Design support conditions at the pier and abutments were
idealized, and a self-weight function was used to apply all dead loads.
Nodes were placed at all cross-frame and flange transitions along the girder lines. Girders were modeled as
straight frame members between these nodes, following common grillage modeling practice. All flange and
web transitions were considered, as well as radial and tangential elevation changes. Intermediate stiffeners
were not considered by the grillage model. K-shaped cross-frames were modeled as straight frame
members and assigned equivalent stiffness values, again, following common grillage modeling practice.
Geometric properties required for a typical cross-frame member were established by modeling a typical
cross-frame and included strong and weak axis moments of inertia, vertical and horizontal shear areas,
torsional constants, and an equivalent area for self-weight calculations. Material response was assumed to
be elastic.
216
9.3.1.2
Full Shell (3D)
Full shell three-dimensional models used for predicting construction response were also developed in
ABAQUS in association with earlier model validation studies and, as such, are largely described in
previous submittals to PennDOT (Linzell et al. 2006). Therefore, a brief summary is provided herein.
ABAQUS S4R elements were used to model the flanges and webs of all girders with webs shells
distributed in a fashion that maintained an element aspect ratio close to 1:1. Flange nodes corresponded to
web nodes along the length of the girders. Flange element aspect ratios did not exceed 2:1. Radial crossframe members between adjacent girders were assumed rigidly connected to the girder webs and were
modeled using B31OS beam elements, with the diagonals and the top and bottom chords being modeled as
separate members for each cross-frame. Girder web stiffeners were modeled using ABAQUS B31 beam
elements. The slab was modeled using S4R shell elements, with slab elements having an aspect ratio of
2:1. All components of the model were assigned nominal material and geometric properties. Boundary
conditions were input to match what was observed from the design plans, depending on the structure,
during erection. Composite action was maintained for appropriate stages using rigid links. Loads consisted
of bridge component dead loads.
9.3.1.3
Shell and Beam (3D)
Based on report results from earlier studies (discussed in the following sections) and published literature
(Chavel and Earls 2006a, 2006b; Kim et al. 2005), the full shell three-dimensional models were modified
slightly to reduce analysis time and provide an additional analysis approach that could be examined. A
summary of these models, largely focusing on the modifications instituted when compared to the full shell
models, is provided herein.
Girder webs were still modeled using ABAQUS S4R shell elements with webs shells distributed in a
fashion that maintained a 1:1 element aspect ratio. Flange shell elements were replaced with ABAQUS B31
beam elements with flange nodes corresponding web nodes. The B31 beam element has 6 degrees of
freedom and allows for transverse shear deformation. It can accurately model both vertical and lateral
bending effects in the flanges. ABAQUS calculates stress values at 25 integration points within the B31
elements. Figure 238 shows the locations of the 25 integration points within a typical flange cross-section.
Flanges were not offset from the girder web initially because of the large depth of the web 2743 mm (108
in) compared to the small average flange offset distance of 60 mm (2.3 in). The effect of this decision was
examined and is discussed in the following sections. The stress values at the tips of the flanges can be used
to calculate both the vertical and lateral bending of the flanges. Radial cross-frame members between
adjacent girders and web stiffeners were still modeled using B31OS elements. The slab was still modeled
using S4R shell elements with an aspect ratio of 2:1. Nominal material and geometric properties were again
used, and boundary conditions matched design plans or observed erection conditions. Composite action
was again maintained using rigid links. Loads again consisted of component dead loads. A typical crosssection with modeling details is shown in Figure 239.
217
Figure 238. Representative B31 Flange Element.
Figure 239. Typical Cross-Section, Shell and Beam Model Construction.
9.3.1.4
Frame, Shell, Brick (3D)
Three-dimensional models containing a combination of frame, shell and brick elements and used for
predicting construction response were also developed in ABAQUS in association with earlier model
validation studies. Therefore, these models are largely described in previous submittals to PennDOT
(Linzell et al. 2006), and a brief summary of these models, largely focusing on the modifications when
compared to the full shell models, is provided herein.
Girder webs were, again, still modeled using ABAQUS S4R shell elements. Flanges were modeled using
B31 elements with cross-frame members between adjacent girders and web stiffeners modeled using
B31OS elements. The slab was modeled using ABAQUS C3D8R brick elements, which are eight-noded,
reduced integration, linear brick elements. Composite action was maintained using rigid links. Nominal
material and geometric properties were used with boundary conditions matching design plans or observed
conditions. Loads consisted of component dead loads.
9.3.2
Analysis Techniques
As discussed in a previous submittal to PennDOT (Linzell et al. 2006), irrespective of the software used to
complete the analysis and the type of model being examined, analyses proceeded using a number of stages
218
that applied the dead load of the structure in sequential steps based on the construction procedure. During
each stage, elements corresponding to the components that were added or placed for that stage were added
to the model’s deformed shape, and the next analysis step was performed. All relevant information was
retained from each stage and used as initial conditions for the next construction step.
9.4
Comparisons and Modifications
Previous submittals to PennDOT (Linzell et al. 2006) included comparisons between field data recorded
during construction of Structure #207 and results from the (1) grillage, (2) full shell, and (3) frame, shell,
and brick models. The reported information focused on stress change predictions during the construction
process for this structure and, based upon these comparisons, were largely inconclusive regarding a
preferred modeling technique. All models were shown to effectively predict trends recorded during
construction, but accuracy varied for predicting stress changes at known locations for specific construction
events. Prediction accuracy improved for events that occurred after slab placement was initiated, but, again,
no definitive conclusions could be made regarding a preferred modeling scheme. Since no definitive
conclusions were obtained from these studies, additional comparisons and modifications were completed to
obtain clearer information regarding selecting a preferred analysis approach and software package for
future tasks.
9.4.1
Comparisons – Structure #207
Primary comparisons were performed between the field data recorded during construction of Structure
#207, the grillage model, and the shell and beam model. The shell and beam model was selected for
comparison since it offered similar levels of accuracy to the other 3D models and improved computational
efficiency.
Comparisons were completed for the construction stages listed in Table 36. Figure 240 shows a
representative erection drawing for an early stage of construction. For the shell and beam model, the
ABAQUS MODEL CHANGE command was used to assign zero stiffness to the elements that were not
present during certain stages of construction. For example, for the first stage of the analysis, only the
elements corresponding to Girders 4 and 5 from Abutment 1 to Field Splice 3 (black lines Figure 240) were
capable of carrying load. The remaining elements in the model (gray lines in Figure 240) were assigned a
zero stiffness value so that they did not influence sequential analysis results. The shell and beam model
was initially run assuming small strain and small deformation response to establish a baseline of its
performance.
Table 36. Structure #207 Examined Construction Stages.
Stage Number
Section Placed During Stage*
Stage 1
Girders 4 and 5 Abutment 1 to Field Splice 3
Stage 2
Girder 3 Abutment 1 to Field Splice 3
Stage 4
Girders 1 and 2 Abutment 1 to Field Splice 3
Stage 5
Girders 4 and 5 Field Splice 3 to Abutment 2
Stage 6
Girder 3 Field Splice 3 to Abutment 2
Stage 8
Girders 1 and 2 Field Splice 3 to Abutment 2
Stage 10
First Portion of Slab
Stage 11
Second Portion of Slab
Stage 12
Third and Final Portion of Slab
*Includes all relevant cross-frames and abutments
219
SHORING TOWER
SHORING TOWERS
ABU
T2
FS 4
G4 C
G4B
G4A
FS 3
PI ER
FS 2
FS 1
ABUT 1
Figure 240. Structure #207 Construction Stage.
To establish a more definitive assessment of model accuracy when compared to field data comparisons
during placement of the steel superstructure, examination of vertical and lateral bending moment
distribution at locations corresponding to instrumented sections on Structure #207 was done. Examining
moment distribution at select sections provides a more realistic picture of model accuracy for the entire
structure at the completion of construction events when compared to comparison of single stress values at
specific locations. During placement of the concrete deck, however, stresses at specific instrument
locations had to again be examined because a number of girder top flange gauges on Structure #207 were
damaged during the deck pour. In addition to the moment distribution and stress comparisons, vertical
deformations predicted by the models were compared to those measured in the field.
9.4.1.1
Vertical Bending Moment Distribution – Superstructure Erection
Vertical bending moment distribution values from the computer models were calculated by summing all the
girder vertical bending moments across a cross-section of the structure and then dividing each individual
girder vertical bending moment value by the sum. A similar approach was used for the field data except
that data started with strains at specific points rather than stresses. Methods for establishing moments for
the computer and field data were outlined in previous submittals to PennDOT (Linzell et al. 2003; Linzell
et al. 2006).
Representative vertical bending moment distribution plots will be presented and discussed herein. Results
refer to stage numbers from Table 36, and sections refer to Figure 241. Representative calculations are
provided in Appendix A.
220
Figure 241. Structure #207 Instrumented Sections.
221
G5
G4
G3
G2
G1
VARIES
G1 - 48'-4"
G5 - 46'-10" A
A
WEST ABUTMENT
F.S. 1
B
B
N
F.S. 3
S
VARIE 5"
'G1 - 96 '-11"
2
G5 - 9
C
C
D
EAST ABUTMENT
D
GIRDER VW STRAIN GAUGE - SCHEME A (DETAIL 1)
GIRDER VW STRAIN GAUGE - SCHEME B (DETAIL 2)
CROSS FRAME VW STRAIN GAGE (DETAIL 3 & 4)
VW TILTMETER (DETAIL 5)
F.S. 4
BRIDGE S-207 - INSTRUMENT LOCATION PLAN
VARIES
G1 - 56'-11"
G5 - 56'-9"
F.S. 2
PIER
Figure 242 shows the vertical bending moment distribution results for Stage 6 Section B-B. Results for
Girders 2 and 3 are not presented during this stage of construction because one or more of the instruments
located on the Girder 2 and 3 flanges were malfunctioning during field testing. The figure shows that the
ABAQUS three-dimensional model over-predicts the vertical bending moment distribution to Girders 1 and
5 when compared to field test results. The figure also shows that the ABAQUS three-dimensional finite
element model predicts a smaller vertical bending moment distribution value for Girder 4 when compared
to the field results. Figure 242 shows that the SAP2000 grillage model over-predicts the vertical bending
moment distribution observed in the field for Girder 1 and under-predicts the vertical bending moment
distribution observed in the field for Girders 4 and 5.
0.6
0.4
No Field Data Available
0.3
Moment Distribution
0.5
G3 B-B
G2 B-B
0.2
0.1
0
G5 B-B
G4 B-B
G1 B-B
Location
FIELD DATA
ABAQUS 3D
SAP2000 GRILLAGE
Figure 242. Stage 6 Vertical Bending Moment Distribution for Section B-B, Structure #207.
Figure 243 shows the vertical bending moment distribution comparisons for Stage 8b. Stage 8b represents
the point in time when the entire steel superstructure had been erected and all the bolts had been tightened.
The figure shows that the ABAQUS three-dimensional finite element model predicts a vertical bending
moment distribution that is very similar to results obtained from the field. The SAP2000 grillage model did
not predict vertical bending moment distribution as accurately.
222
0.45
0.4
Moment Distribution
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
G5 B-B
G4 B-B
G3 B-B
G2 B-B
G1 B-B
Location
FIELD DATA
ABAQUS 3D
SAP2000 GRLLAGE
Figure 243. Stage 8b Vertical Bending Moment Distribution for Section B-B, Structure #207.
Figure 244 shows the vertical bending moment distribution results for Stage 5 Section C-C. Results are not
presented for Girders 1, 2, and 3 because these sections are not in place during this stage of construction. It
shows that the SAP2000 grillage model predicts the vertical bending moment distribution field behavior
slightly better than the ABAQUS three-dimensional finite element model.
223
Moment Distribution
0.5
0.4
0.3
0.2
0.1
0
G5 C-C
G4 C-C
G3 C-C
G2 C-C
Section Not in Place During This Phase of Construction
0.6
0.7
G1 C-C
Location
FIELD DATA
ABAQUS 3D
SAP2000 GRILLAGE
Figure 244. Stage 5 Vertical Bending Moment Distribution for Section C-C, Structure #207.
Figure 245 shows the vertical bending moment distribution results for Stage 6 Section C-C. Results are not
presented for Girders 1 and 2 because these sections are not in place during this stage of construction. The
ABAQUS three-dimensional finite element model over-predicts the vertical bending moment distribution
values for Girders 4 and 5 when compared to the field test values and predicts smaller vertical bending
moment distribution values for Girder 3 when compared to the field test results. The figure shows that the
SAP2000 grillage model predicts a similar vertical bending moment distribution value for Girder 5 when
compared to the field test results but that it over-predicts the vertical bending moment distribution value for
Girder 4 and under-predicts the model vertical bending moment distribution value for Girder 3.
224
Moment Distribution
0.5
0.4
0.3
0.2
0.1
0
G5 C-C
G4 C-C
G3 C-C
G2 C-C
Section Not in Place During This Phase of Constructon
0.6
G1 C-C
Location
FIELD DATA
ABAQUS 3D
SAP2000 GRILLAGE
Figure 245. Stage 6 Vertical Bending Moment Distribution for Section C-C, Structure #207.
Figure 246 shows that neither the ABAQUS three-dimensional finite element model nor the SAP2000
grillage model predicts the vertical bending moment distribution values observed in the field for Stage 8b
Section C-C. However, the ABAQUS three-dimensional finite element model predicts the general vertical
bending moment distribution behavior observed in the field throughout the entire cross-section more
accurately than the SAP2000 model.
The vertical bending moment distribution comparisons between the field data, the ABAQUS threedimensional finite element model, and the SAP2000 grillage model provide conclusive evidence that the
ABAQUS three-dimensional finite element model is more accurate than the SAP2000 grillage model in
predicting the field vertical bending moment distribution behavior of the steel superstructure during
erection.
225
0.35
0.3
Moment Distribution
0.25
0.2
0.15
0.1
0.05
0
G5 C-C
G4 C-C
G3 C-C
G2 C-C
G1 C-C
Location
FIELD DATA
ABAQUS 3D
SAP2000 GRILLAGE
Figure 246. Stage 8b Vertical Bending Moment Distribution for Section C-C, Structure #207.
When the vertical bending moment distribution comparisons are examined to ascertain a preferred
modeling method, the results can be interpreted as mixed. However, they do show an improvement for the
ABAQUS 3D model when compared to the SAP2000 grillage model for predicting general distribution
trends.
9.4.1.2
Lateral Bending Moment Distribution – Superstructure Erection
Lateral bending moment distributions were calculated using the same procedure that was followed for the
vertical bending moment distribution calculations. Representative lateral bending moment distribution plots
will be presented and discussed herein. Results refer to stage numbers from Table 36, and sections refer to
Figure 241. Representative calculations are provided in Appendix A.
Figure 247 shows the lateral bending moment distribution for Stage 8b. The figure shows that neither the
ABAQUS three-dimensional finite element nor the SAP2000 grillage model predicts the field lateral
bending moment distribution for Section B-B well. The ABAQUS three-dimensional finite element model
predicts similar lateral bending distribution values for Girders 1 and 4 but predicts significantly different
values for G3 and G5. The figure also shows that the SAP2000 model predicts significantly different
lateral moment distribution values for Girder 1, 3, 4, and 5 when compared to the field data.
226
0.5
0.45
0.4
0.3
0.25
0.2
Moment Distribution
0.35
0.15
0.1
0.05
0
G5 B-B
G4 B-B
G3 B-B
G2 B-B
G1 B-B
Location
FIELD DATA
ABAQUS 3D
SAP2000 GRILLAGE
Figure 247. Stage 8b Lateral Bending Moment Distribution for Section B-B, Structure #207.
Figure 248 shows the lateral bending moment distribution results for Section C-C during Stage 8b of
construction. The figure shows that the ABAQUS three-dimensional finite element model and the
SAP2000 grillage model again do not predict the field lateral bending moment distribution accurately.
There appears to be no pattern for discrepancies between the field lateral bending moment distribution
values and those from the ABAQUS and the SAP2000 models.
227
0.5
0.45
0.4
Moment Distribution
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
G5 C-C
G4 C-C
G3 C-C
G2 C-C
G1 C-C
Location
FIELD DATA
ABAQUS 3D
SAP2000 GRILLAGE
Figure 248. Stage 8b Lateral Bending Moment Distribution for Section C-C, Structure #207.
The figures above are representative plots of lateral bending moment distribution comparisons for the other
stages of construction. These comparisons do not clearly identify which model more effectively predicts
measured lateral bending moment behavior; in fact, neither the ABAQUS model nor the SAP2000 model
effectively predicted lateral bending moment distributions observed in the field. However, it should be
emphasized that lateral bending stresses for most construction stages were small compared to vertical
bending stresses for those stages, and, as such, ineffective prediction of individual lateral bending stresses
or the distribution of lateral bending stresses at a bridge cross-section cannot be construed as conclusive
evidence that a modeling approach is ineffective.
9.4.1.3
Vertical Bending Stresses – Deck Pour
As stated earlier, comparisons during the deck pour had to be made by examining stress measurements at
individual points because during the field test, a majority of the top flange strain gauges were damaged
during the placement of the deck pans. Figure 249 shows the sequence of the deck placement used during
construction. Stress comparisons will be presented for Deck Pour 1 (Stage 10), Deck Pour 2 (Stage 11),
and Deck Pour 3 (Stage 12). Representative calculations are provided in Appendix A.
228
UT)
CKO
" BLO
' -6
m (3
4
1.07
1/2")
'- 4
m (1
60.92
34.10 m (111' -10 1/2")
1.07 m (3'-6" BLOCKOUT)
4
0.42 m (1' -4 1/2")
1
C
L A BUTMENT #1
63.56 m (208'-6")
SPA N 1
3
")
49.38 m (162'-0
CL P IER
'-6")
m (269
82.13
2
S PA N
2
' -1
m (199
0 1/2")
0.42
C
L A BUTMENT #2
Figure 249. Deck Placement Sequence, Structure #207.
Figure 250 shows stress comparisons for Girder 1 bottom flange Section A-A. It shows that the ABAQUS
three-dimensional model predicts smaller stress values than those observed in the field for all three stages
229
of the deck placement. The SAP2000 grillage model also under-predicts the stress values observed in the
field for Girder 1 Section A-A for all three stages of deck placement.
16.0
100.0
Vertical Bending Stress (ksi)
12.0
80.0
10.0
60.0
8.0
6.0
40.0
Vertical Bending Stress (MPa)
14.0
4.0
20.0
2.0
0.0
0.0
10 - Deck Pour 1
11 - Deck Pour 2
12 - Deck Pour 3
Event
SAP2000 GRILLAGE
FIELD DATA
ABAQUS 3D
Figure 250. Stress Comparisons for Girder 1 Bottom Flange Section A-A, Structure #207.
Figure 251 shows stress comparisons for the bottom flange of Girder 1 Section B-B during placement of
the concrete deck. It shows that the ABAQUS three-dimensional finite element model predicts stress very
similar to those observed in the field for Girder 1, and it also shows that the SAP2000 model predicts larger
stress values than those observed in the field for Girder 1 Section B-B.
230
0.0
-13.1
-33.1
-53.1
-8.0
-73.1
-12.0
-93.1
-16.0
-113.1
-133.1
-20.0
-4.0
-153.1
-24.0
-173.1
-28.0
-193.1
10 - Deck Pour 1
11 - Deck Pour 2
SAP2000 GRILLAGE
Event
FIELD DATA
12 - Deck Pour 3
ABAQUS 3D
Figure 251. Stress Comparisons for Girder 1 Bottom Flange Section B-B, Structure #207.
Figure 252 shows stress comparisons for the top flange of Girder 3 Section B-B. It shows that the
ABAQUS three-dimensional finite element model predicts stress values smaller than those observed in the
field for the top flange of Girder 3 Section B-B, and the SAP2000 grillage model also predicts stress values
smaller than those observed in the field for the top flange of Girder 3 Section B-B.
231
25.0
160.0
120.0
15.0
100.0
80.0
10.0
60.0
140.0
20.0
40.0
5.0
20.0
0.0
0.0
10 - Deck Pour 1
11 - Deck Pour 2
12 - Deck Pour 3
Event
SAP2000 GRILLAGE
FIELD DATA
ABAQUS 3D
Figure 252. Stress Comparisons for Girder 3 Top Flange Section B-B, Structure #207.
The previously discussed figures show that for certain girder locations, both the ABAQUS threedimensional finite element model and the SAP2000 grillage model accurately predict stress values
observed in the field. However, for other locations, the predictions are not as accurate. These mixed
results match those for earlier comparisons involving stress values at a single point and, as discussed earlier,
cannot be construed as definitive information that can be used to select or reject one modeling approach
over another. Since no conclusive evidence is provided using these comparisons, similar to what was found
with earlier single-point stress comparisons, another global performance measure, such as structure
deformations or moment distribution at a given cross-section as presented earlier, would be needed to assist
with making a definitive model selection decision. Vertical deflections were selected as the global
parameter to examine since extensive field data were available for the entire structure using a combination
of tiltmeters and laser measurements. The next section contains deflection comparisons between the
ABAQUS three-dimensional finite element model, the SAP2000 grillage model, and the field data.
9.4.1.4
Vertical Deflections – Deck Pour
This section contains deflection comparisons between the ABAQUS three-dimensional finite element
model, the field data, and the SAP2000 grillage model. Deflection data obtained from the field test
encompassed changes in deformations for all girders between erection of the entire steel superstructure and
placement of the slab. Values from the computer models were obtained using sequential analyses that
followed the same construction sequence. Figure 253 through Figure 257 show deflections for Girders 1
through 5, respectively.
232
20
40
60
80
100
120
140
2.0
46.0
0.0
-4.0
-2.0
-54.0
-4.0
-104.0
-6.0
-154.0
-8.0
-204.0
-10.0
0
50
100
150
200
250
300
350
400
450
-254.0
500
FIELD DATA
SAP2000 GRILLAGE
ABAQUS 3D
Superstructure and Placement of the Slab, Structure #207.
233
Deflection, mm
Deflection, in.
0
20
40
60
80
100
120
140
2.0
46.0
0.0
-4.0
-2.0
-54.0
-4.0
-104.0
-6.0
-154.0
-8.0
-204.0
-10.0
0
50
100
150
200
250
300
350
400
450
-254.0
500
FIELD DATA
SAP2000 GRILLAGE
ABAQUS 3D
234
Deflection, mm
Deflection, in.
0
20
40
60
80
100
120
140
2.0
46.0
0.0
-4.0
-2.0
-54.0
-4.0
-104.0
-6.0
-154.0
-8.0
-204.0
-10.0
0
50
100
150
200
250
300
350
400
450
-254.0
500
FIELD DATA
SAP2000 GRILLAGE
ABAQUS 3D
235
Deflection, mm
Deflection, in.
0
0
20
40
60
80
100
120
140
1.0
0.0
-4.0
-1.0
-54.0
-3.0
-4.0
-104.0
-5.0
-6.0
-154.0
-7.0
-8.0
-204.0
-9.0
-10.0
0
50
100
150
200
250
300
350
400
450
-254.0
500
FIELD DATA
SAP2000 GRILLAGE
ABAQUS 3D
236
Deflection, mm
Deflection, in.
-2.0
0
20
40
60
80
100
120
140
1.0
21.4
0.0
-1.0
-28.6
-3.0
-78.6
-4.0
-5.0
-128.6
Deflection, mm
Deflection, in.
-2.0
-6.0
-7.0
-178.6
-8.0
-9.0
0
50
100
150
200
250
300
350
400
450
-228.6
500
FIELD DATA
SAP2000 GRILLAGE
ABAQUS 3D
The figures show that both the SAP2000 grillage model and the ABAQUS three-dimensional finite element
model predict the general behavior of girder vertical deflections observed in the field. However, the
ABAQUS three-dimensional finite element model predicts the magnitude of the deflections of all of the
girders more accurately than the SAP2000 grillage model when compared to field results. As a result of
these comparisons, the ABAQUS shell and beam models were tentatively selected for use with future
parametric studies.
9.4.2
Modifications
It was of interest to examine whether additional modifications to the selected modeling technique improved
model accuracy. As discussed in previous submittal to PennDOT (Linzell et al. 2003; Linzell et al. 2006),
earlier studies investigated the effects of certain parameter variations, such as boundary conditions, on
results. Additional modifications that were explored included: (1) examining the effectiveness with which
first-order displacement approaches selected for the previous comparisons predicted response for beams
with greater horizontal curvature effects, (2) studying how changes in the sequential slab placement
procedure used for the computer model affected results, and (3) examining how changes to the offset
between the girder web and flanges elements affected results.
9.4.2.1
First-Order Displacements
The effectiveness of first-order displacement assumptions for predicting the response of beams having
higher horizontal curvature influence was assessed using comparisons to experimental data from a study by
Heins and Spates (1968). They conducted tests of a horizontally curved steel beam with a web depth of
177.8mm (7.0 in) and a thickness of 6.35mm (0.25 in). The flanges were 93.0 mm (3.7 in) by 10.0 mm (0.4
in). The beam was 9.1m (30.0 ft) long and had a radius of curvature of 15.2 m (50.0 ft). The ends of the
girder were encased in concrete blocks to replicate fixed end conditions. The concrete blocks encased 0.5
m (1.5 ft) of each end of the girder, resulting in a clear span of 8.2 m (27.0 ft). The R/L ratio of the beam
237
was 1.85, indicating extreme curvature effects. The beam was tested under either a concentrated vertical
load at midspan or a concentrated vertical load at its three-tenths point. The load in each case was 4448 N
(1000 lb). It was instrumented with gauges to record deformations while under the test loads.
A three-dimensional finite element model of the test beam was created in ABAQUS following the shell and
beam technique. The web was modeled using 7 S4R shell elements, with nodes along the girders placed at
approximately 25 mm (1 in) intervals to keep the element aspect ratio as close to 1 as possible. The flanges
were modeled using beam elements. Nodes along the ends of the beam were restrained from all movement
to model the fixed support conditions. The beam was analyzed under two loading conditions: (1)
concentrated load at midspan and (2) concentrated load at the three-tenths point.
Figure 258 and Figure 259 show representative deformation comparisons between test and computer model
results. These comparisons indicate that using a first-order displacement approach does an effective job of
predicting deformations, including rotations, for a beam with severe horizontal curvature. While rotations
are not predicted with as much accuracy as vertical deflections, the fact that they are over-predicted by the
model coupled with the small magnitude of the rotations that were measured supports using a first-order
approach for future parametric studies.
Distance from End of Beam, m
0
1
2
3
4
5
6
7
8
9
0.00
-0.70
-0.05
-0.10
-2.70
-4.70
-0.20
-0.25
-6.70
-0.30
-8.70
-0.35
-0.40
-10.70
-0.45
-0.50
-12.70
0
5
10
15
20
25
30
Distance from End of Beam, ft.
ABAQUS
Heins Spates Experimental
Figure 258. Vertical Deflection Comparison, Concentrated Load at Midspan, Heins and Spates
(1968) Tests.
238
Deflection, mm
Deflection, in.
-0.15
Distance from End of Beam, m
0
1
2
3
4
5
6
7
8
9
0.03
0.02
Rotation, radians
0.01
0.00
-0.01
-0.02
-0.03
-0.04
-0.05
0
5
10
15
20
25
30
Distance from End of Beam, ft.
ABAQUS
Heins Spates Experimental
Figure 259. Rotation Comparison, Concentrated Load at Three-Tenths Point, Heins and Spates
(1968) Tests.
9.4.2.2
Slab Placement
A modified technique for numerically placing the concrete slab was examined to determine its influence on
results. The initial approach for adding deck concrete to the model involved placing each slab section in
one step, thereby assuming that the concrete had achieved its full stiffness at the time of placement. To
examine the influence of fresh concrete on computer results models were created that incorporated an
additional step for each slab section that included slab shell elements with full weight but negligible
stiffness. To simulate setting of the concrete a subsequent step that removed the ―wet‖ concrete shells from
the active list of elements in the model and added shells that had the full slab stiffness to the list of active
elements.
To assess the influence of changing deck placement techniques on model results, final deformations
predicted by each of the techniques were compared for Structure #207. Figure 260 and Figure 261 show
representative final deflection results for Girder 1 for the original slab placement method (only one slab
step per section placed) and the modified slab placement method (two steps per slab section placed). Only
minor differences are shown, and as a result, the original technique was retained for future models due to a
combination of its effectiveness and modeling efficiency.
239
20
40
60
80
100
120
140
2.0
44.4
0.0
-5.6
-2.0
-55.6
-4.0
-105.6
-6.0
-155.6
-8.0
-205.6
-10.0
-255.6
-12.0
-305.6
-14.0
0
50
100
150
200
250
300
350
400
450
Deflection, mm
Deflection, in.
0
-355.6
500
G1 Orginal Slab
G1 Modified Slab
Figure 260. Vertical Deflection Comparisons, Original and Modified Slab Placement Techniques,
Structure #207.
240
20
40
60
80
100
120
140
10.0
247.6
8.0
197.6
6.0
147.6
4.0
97.6
2.0
47.6
0.0
-2.4
-2.0
-52.4
-4.0
-102.4
-6.0
0
50
100
150
200
250
300
350
400
450
Deflection, mm
Deflection, in.
0
-152.4
500
G1 Orginal Slab
G1 Modified Slab
Figure 261. Radial Deflection Comparisons, Original and Modified Slab Placement Techniques,
Structure #207.
9.4.2.3
Flange Offset
The original shell and beam models did not include flange offsets between the web shells and flange beam
elements. A second model was created that incorporated flange nodes that were displaced a distance equal
to half the thickness of the flange from the web shell nodes. Comparisons again occurred for Structure
#207 and, in this case, improvement offered by including the offset via revised moment distribution
comparisons during steel erection between the original ABAQUS finite element model, the ABAQUS
model using flange offsets, the SAP2000 grillage model, and the field data were examined. A single
representative comparison is shown in Figure 262, and it indicates that including flange offsets did not
appreciably improve model accuracy. These results and those from earlier sections indicate that initial
modeling assumptions for the shell and beam models were acceptable for the parametric studies.
241
0.35
0.3
Moment Distribution
0.25
0.2
0.15
0.1
0.05
0
G5 C-C
G4 C-C
G3 C-C
G2 C-C
G1 C-C
Location
FIELD DATA
ABAQUS 3D
SAP2000 GRILLAGE
ABAQUS FLANGE OFFSET
Figure 262. Stage 8b Vertical Bending Moment Distribution for Section C-C, Including Flange
Offset, Structure #207.
9.5
Additional Evaluations
After recommendation of a modeling technique for future project tasks based on the comparisons and
modification examination outlined previously, additional evaluation of the recommended technique
occurred via comparisons to data recorded by Penn State personnel from other curved and skewed
structures during construction. These structures included two curved plate girder bridges, Structure #7A
and the I-79 Missing Ramps Bridge, and a single skewed plate girder structure, Structure #28. Evaluations
that were completed and corresponding discussions are provided in the sections that follow.
9.5.1
Evaluations – Curved Bridges
As discussed in previous submittals to PennDOT (Linzell et al. 2003) and in previous sections, data for
evaluating the effectiveness of the recommended model for predicting the response of Structure #7A and
the Missing Ramps Bridge was collected during specific instances of the erection process, and, as such,
limited comparisons are completed and discussed herein.
9.5.1.1
Structure #7A
Construction of the ABAQUS shell and beam model for Structure #7A followed the procedure discussed in
previous sections. A view of the superstructure model, detailing element types and levels of discretization,
is shown in Figure 263.
242
Figure 263. ABAQUS Model, Structure #7A.
As was discussed in earlier submittals to PennDOT (Linzell et al. 2003), deformation measurement surveys
were taken during the realignment procedure by a third party. Representative comparisons between
deformations recorded during these surveys and predictions from the ABAQUS models are shown in
Figure 264 through Figure 268. These comparisons detail changes in vertical displacements during
realignment of the superstructure in Spans 4 and 5 and indicate good agreement between measured and
predicted values.
243
Distance from Abut., m
0
20
40
60
80
100
120
0.6
0.4
117.1
0.2
67.1
17.1
0.0
-32.9
-0.2
Vertical Deflection, mm
Vertical Deflection, ft.
167.1
-82.9
-0.4
-132.9
-0.6
-182.9
0
50
100
150
200
250
300
350
400
Distance from Abut., ft.
ABAQUS
FIELD DATA
Figure 264. G1Vertical Displacement Changes During Realignment of Spans 4 and 5, Structure #7A.
0
20
40
60
80
100
120
0.50
128.08
0.40
0.30
0.20
0.10
28.08
0.00
-21.92
-0.10
78.08
-0.20
-71.92
-0.30
-0.40
-121.92
0
50
100
150
200
250
300
350
400
ABAQUS
FIELD DATA
244
0
20
40
60
80
100
120
0.40
108.56
0.20
58.56
0.10
8.56
0.00
-0.10
0.30
-41.44
-0.20
-0.30
-91.44
0
50
100
150
200
250
300
350
400
ABAQUS
FIELD DATA
245
0
20
40
60
80
100
120
0.30
89.52
0.25
69.52
49.52
0.15
0.10
29.52
0.05
9.52
0.20
0.00
-10.48
-0.05
-0.10
-30.48
0
50
100
150
200
250
300
350
400
ABAQUS
FIELD DATA
0
20
40
60
80
100
120
0.25
74.76
64.76
0.20
0.15
44.76
34.76
0.10
24.76
0.05
14.76
54.76
4.76
0.00
-5.24
-0.05
-15.24
0
50
100
150
200
250
300
350
400
ABAQUS
FIELD DATA
Figure 268. G5 Vertical Displacement Changes During Realignment of Spans 4 and 5, Structure #7A.
246
9.5.1.2
Missing Ramps Bridge
Construction of the ABAQUS shell and beam model for the Missing Ramps Bridge followed the procedure
outlined and evaluated in the previous sections. A view of the steel superstructure model is shown in
Figure 269. Figure 270 details cross-frame labels used for model evaluations.
Figure 269. ABAQUS Model, Missing Ramps Bridge.
Figure 270. ABAQUS Model Detailing Cross-Frame Labels, Missing Ramps Bridge.
As was discussed in previous sections, PennDOT personnel measured girder rotations at cross-frame
connection locations during construction of the steel superstructure by measuring radial translations at the
web-top flange junction. Figure 271 through Figure 273 present representative comparisons between
predicted and measured girder rotations at the completion of erection at select cross-frame locations.
Comparisons are presented for Unit 1, followed by Unit 2. These comparisons show good agreement
between predicted and measured lateral displacements at most girder locations and quite good prediction of
girder movements at radial cross-sections through the bridge.
247
Comparison between Field Data and 3D ABAQUS model at
completion of girder erection (Cross-Frame 1)
1.00
0.90
Field Data
Radial Deflection, in
0.80
3D ABAQUS Model
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
G1
G2
G3
G4
G5
Girder Position
Figure 40. Girder Radial Translations at Web-Flange Junction, Cross-Frame 1, Completion of Unit 1
Girder Erection, Missing Ramps Bridge.
1.00
0.90
Field Data
0.80
3D ABAQUS Model
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
G1
G2
G3
G4
G5
Girder Position
Figure 271. Girder Radial Translations at Web-Flange Junction, Cross-Frame 10, Completion of
Unit 1 Girder Erection, Missing Ramps Bridge.
248
1.00
0.80
Field Data
3D ABAQUS Model
0.60
0.40
0.20
0.00
-0.20
G1
G2
G3
G4
G5
-0.40
-0.60
-0.80
-1.00
Girder Position
249
1.00
0.80
Field Data
0.60
3D ABAQUS Model
0.40
0.20
0.00
-0.20
G1
G2
G3
G4
G5
-0.40
-0.60
-0.80
-1.00
Girder Position
9.5.2
Evaluations – Skewed Bridge
As discussed in previous submittals to PennDOT, limited data that encompassed response of the structure
during deck placement were collected for Structure #28. As a result, evaluations of model effectiveness
consisted of examination of deformation predictions as the concrete deck was being placed.
9.5.2.1
Structure #28
Construction of the ABAQUS shell and beam model for Structure #28 followed the procedure outlined and
evaluated in the previous sections. A view of the steel superstructure model is shown in Figure 274.
250
Figure 274. ABAQUS Model, Structure #28.
As was discussed in previous submittals to PennDOT (Linzell et al. 2003), the majority of the
measurements taken during the deck pour were deformations. One set of deformation measurements
involved tracking girder rotation at the abutments by measuring out-of-plane girder web displacement at its
intersection with the top flange. Figure 275 details representative differences between final displacements
at one (east) end of select girders and shows the effectiveness of the model for predicting those values.
One additional set of measurements that was taken involved girder vertical displacements, and Figure 276
and Figure 277 show representative results for fascia girders G1 and G7. Again, the effectiveness of the
model for predicting deformations is demonstrated.
1.00
Field Data
Horizontal Displacements, in
0.90
3D ABAQUS Model
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
G2-East
G4-East
G6-East
Girder Location
Figure 275. Girder Lateral Displacements, Completion of Deck Pour, Structure #28.
251
Comparison between Field Data and 3D ABAQUS model results
at completion of concrete pour (G1)
0
0
50
100
150
200
250
300
Vertical Deflection, ft
-0.1
-0.2
-0.3
-0.4
Field Data
3D ABAQUS Model
-0.5
-0.6
Distance from West Abutment, ft
Figure 276. G1 Vertical Displacements, Completion of Deck Pour, Structure #28.
Comparison between Field Data and 3D ABAQUS model results
at completion of concrete pour (G7)
0
0
50
100
150
200
250
Vertical Deflection, ft
-0.1
-0.2
-0.3
-0.4
-0.5
Field Data
3D ABAQUS Model
-0.6
-0.7
Distance from West Abutment, ft
Figure 277. G7 Vertical Displacements, Completion of Deck Pour, Structure #28.
252
300
9.6
Conclusions
Summarized herein were modeling development and verification steps completed in association with Work
Order 009, Guidelines for Analyzing Curved and Skewed Bridges and Designing Them for Construction.
Contained within this report were sections that provided information on structures that were modeled,
discussed modeling techniques that were examined, and explained comparisons, modifications, and
evaluations that occurred to assist with selecting a model type to be used to complete future project tasks.
As a result of the work, a shell and beam model is recommended for future parametric studies. This model
utilizes ABAQUS S4R shell elements for the girder webs with no offsets between the web and girder top
and bottom flanges. ABAQUS B31 beam elements are used for girder flanges with cross-frame members
and web stiffeners modeled using B31OS elements. The model is analyzed using a first-order deformation
approach. When the deck pour is modeled, it is applied to the model using a single group of shell elements
having hardened deck properties.
This type of model was selected because of:
the computational efficiency and flexibility offered by an ABAQUS shell and beam model when
compared to grillage models and three-dimensional finite element models involving purely shell
elements to model the girder or a combination of shell and solid elements,
its effectiveness for predicting deformations, variables that are of primary interest during erection,
for various types of structures and at various instances during construction, and
its similar to slightly improved level of accuracy for predicting bending moment distributions
within curved structures when compared to grillage models or other three-dimensional models.
253
9.7
References
Chavel, B. W., and Earls, C. J. (2006a). ―Construction of a Horizontally Curved Steel I-Girder Bridge. Part
I: Erection Sequence.‖ J. Bridge Eng., 11(1), 81-90.
Chavel, B. W., and Earls, C. J. (2006b). ―Construction of a Horizontally Curved Steel I-Girder Bridge. Part
II: Inconsistent Detailing.‖ J. Bridge Eng., 11(1), 91-98.
Heins, C.P. and Spates, K.R. (1970), Behavior of Single Horizontally Curved Girder, Journal of the
Structural Division, ASCE, 96(ST7), 1511-1529, USA.
Hiltunen, D.R., Johnson, P.A., Laman, J.A., Linzell, D.G., Miller, A.C., Niezgoda, S.L., Scanlon, A.,
Schokker, A.J. and Tikalsky, P.J. (2004), Interstate 99 Research, Contract No. SPC 020S78, Pennsylvania
Department of Transportation, October, 324 pp.
Kim, Y.D., Jung, S.-K. and White, D.W. (2005), ―Transverse Stiffener Requirements in Straight and
Horizontally Curved Steel I-Girders,‖ ASCE Journal of Bridge Engineering, 12(2), 174-183.
Linzell, D.G., Laman, J.A., Bell, B., Bennett, A., Colon, J., Lobo, J., Norton, E. and Sabuwala, T. (2003),
Prediction of Movement and Stresses in Curved and Skewed Bridges, University-Based Research,
Education and Technology Transfer Program; Agreement No. 359704, Work Order 79. Final Report,
Pennsylvania Department of Transportation, March, 192 pp.
Linzell, D.G., Nadakuditi, V.P. and Nevling, D.L. (2006), Prediction of Movement and Stresses in Curved
and Skewed Bridges, PennDOT/MAUTC Partnership, Work Order No. 2, Research Agreement No.
510401, Final Report, Pennsylvania Department of Transportation, September, 86 pp.
254
Appendix A – Representative Moment and Stress Calculations
Vertical Bending Moment Distribution Representative Calculations
Sample calculations: Use G5 Section B-B Stage 8b (see Table 36, Figure 241). Results shown in BOLD.
Equation 6
where,
ζtf = calculated top flange stress, ksi
ζtffd = field data stress at bottom of top flange, ksi
ζbffd = field data stress at bottom of bottom flange, ksi
ttf = top flange thickness, in.
d = web depth, in.
tbf = bottom flange thickness, in.
Sample Equation 1
Equation 2
where,
M (k-ft) = girder vertical bending moment, k-ft
I = moment of interia, in.4
ybot = distance to the neutral axis from bottom of the girder, in.
Sample Equation 2
Equation 3
where,
MdistG5 = vertical bending moment distribution to Girder 5
MG5 (k-ft) = Girder 5 vertical bending moment, k-ft
∑M (k-ft) = sum of the girder vertical bending moments at radial section, k-ft
Sample Equation 3
The same equations and procedure was used for the ABAQUS and the SAP2000 girder vertical bending
moment.
Lateral Bending Moment Distribution Representative Calculations
Sample calculations: Use G5 Section B-B Stage 8b (see Table 36, Figure 241). Results shown in BOLD.
255
Equation 4
where,
ζvbf = bottom flange vertical bending stress, ksi
ζebf = bottom flange exterior tip vertical bending stress, ksi
ζibf = bottom flange interior tip vertical bending stress, ksi
Sample Equation 4
Equation 5
where,
ML (k- ft) = bottom flange lateral bending moment, k-ft
S = weak axis section modulus of the bottom flange, in.3
Sample Equation 5
Equation 6
where,
MLdistG5 = lateral bending moment distribution to Girder 5
MLG5 (k-ft) = Girder 5 lateral bending moment, k-ft
∑ML (k-ft) = sum of the lateral bending moments at radial section, k-ft
Sample Equation 6
The same equations and procedure was used for the ABAQUS and the SAP2000 girder lateral bending
moment.
Vertical Bending Stresses During Slab Pour Representative Calculations
Sample calculations: Use G3 TF Section B-B Stage 10 (see Figure 249). Results shown in BOLD.
Equation 7
256
where,
µε = microstrain (µε)
B = batch gage factor
R0 = reading from data-logger
Sample Equation 8
Equation 9
where,
ζ = stress, ksi
µε = microstrain (µε)
E = steel elastic modulus, ksi
Sample Equation 10
257
10 APPENDIX C
Parametric Study Bridges
258
APPENDIX C-1
Curved Bridge Drawings
259
260
261
262
263
264
265
266
267
268
269
270
271
APPENDIX C-2
Skewed Bridge Drawings
272
273
274
275
276
277
278
279
280
281
282
283
11 APPENDIX D
PennDOT Documents
284
PENNSYLVANIA
DEPARTMENT
OF
TRANSPORTATION
DESIGN MANUAL
PART 4
VOLUME 1
PART B: DESIGN SPECIFICATIONS
SECTION 2 - GENERAL DESIGN AND LOCATIONS FEATURES
SECTION 2 - TABLE OF CONTENTS
2.3 LOCATION FEATURES .................................................................................................................................................. 1
2.3.2 Bridge Site Arrangement ......................................................................................................................................... 1
2.3.2.2 TRAFFIC SAFETY .......................................................................................................................................... 1
2.3.2.2.1 Protection of Structures .......................................................................................................................... 1
2.3.2.2.2 Protection of Users ................................................................................................................................. ii
2.3.2.2.3 Geometric Standards .............................................................................................................................. ii
2.3.3 Clearances ................................................................................................................................................................. ii
2.3.3.2 HIGHWAY VERTICAL .................................................................................................................................. ii
2.3.3.3 HIGHWAY HORIZONTAL ........................................................................................................................... iii
2.3.3.4 RAILROAD OVERPASS ............................................................................................................................... iii
2.5 DESIGN OBJECTIVES ..................................................................................................................................................... v
2.5.2 Serviceability ............................................................................................................................................................ vi
2.5.2.2 INSPECTABILITY ......................................................................................................................................... vi
2.5.2.2.1P General ................................................................................................................................................ vi
2.5.2.2.2P Inspection Walks ................................................................................................................................. vi
2.5.2.2.2aP General ...................................................................................................................................... vi
2.5.2.2.2bP Design Live Load ...................................................................................................................... vii
2.5.2.2.2cP Geometry ................................................................................................................................... vii
2.5.2.2.2dP Connection to the Main Members ............................................................................................. vii
2.5.2.2.3P Inspectability for Enclosed Section .................................................................................................... vii
2.5.2.2.4P Girder Handrail .................................................................................................................................. vii
2.5.2.2.5P Sound Barriers .................................................................................................................................... vii
2.5.2.6 DEFORMATIONS ......................................................................................................................................... vii
2.5.2.6.1 General ................................................................................................................................................. vii
2.5.2.6.2 Criteria for Deflection ......................................................................................................................... viii
2.5.2.6.3 Criteria for Span-to-Depth Ratios.......................................................................................................... ix
2.5.2.7 CONSIDERATION OF FUTURE WIDENING ............................................................................................. ix
2.5.2.7.1 Exterior Beams on Multi-Beam Bridges ............................................................................................... ix
2.5.3 Constructibility ........................................................................................................................................................ ix
2.5.3.1P FALSEWORK ............................................................................................................................................... x
2.6 HYDROLOGY AND HYDRAULIC ............................................................................................................................... xii
2.6.6 Roadway Drainage.................................................................................................................................................. xii
2.6.6.1 GENERAL ..................................................................................................................................................... xii
2.7P DESIGN DRAWINGS .................................................................................................................................................. xiii
2.7.1P Moment and Shear Envelope Diagrams ............................................................................................................ xiii
2.7.2P Major, Complex and Unusual Bridges .............................................................................................................. xiv
2.7.2.1P LOAD DATA SHEET ................................................................................................................................ xiv
2.8P BRIDGE SECURITY ..................................................................................................................................................... xv
2.8.1P General .................................................................................................................................................................. xv
C3.1 SCOPE .............................................................................................................................................................................. 1
3.3 NOTATION......................................................................................................................................................................... 1
3.3.1 General ...................................................................................................................................................................... 1
3.4 LOAD FACTORS AND COMBINATIONS .................................................................................................................... 1
3.4.1 Load Factors and Load Combinations ................................................................................................................... 1
3.4.1.1P LOAD FACTORS AND COMBINATIONS FOR TYPICAL PENNDOT BRIDGE COMPONENTS ....... 5
3.5 PERMANENT LOADS .................................................................................................................................................... 19
B.2 - cclxxxv
3.5.1 Dead Loads: DC, DW and EV .............................................................................................................................. 19
3.5.1.1P APPLICATION OF DEAD LOAD ON GIRDER AND BOX BEAM STRUCTURES .............................. 19
3.6 LIVE LOADS .................................................................................................................................................................... 19
3.6.1 Gravity Loads: LL, PL .......................................................................................................................................... 19
3.6.1.1 VEHICULAR LIVE LOAD ........................................................................................................................... 19
3.6.1.1.2 Multiple Presence of Live Load ............................................................................................................ 19
3.6.1.2 DESIGN VEHICULAR LIVE LOAD ........................................................................................................... 20
3.6.1.2.1 General ................................................................................................................................................. 20
3.6.1.2.3 Design Tandem..................................................................................................................................... 20
3.6.1.2.5 Tire Contact Area ................................................................................................................................. 20
3.6.1.2.6 Distribution of Wheel Loads through Earth Fills ................................................................................. 21
3.6.1.2.7P Design Permit Load ............................................................................................................................ 21
3.6.1.2.8P Maximum Legal Load (ML-80) ......................................................................................................... 22
3.6.1.3 APPLICATION OF DESIGN VEHICULAR LIVE LOADS ........................................................................ 22
3.6.1.3.1 General ................................................................................................................................................. 22
3.6.1.3.2 Loading for Live Load Deflection Evaluation ...................................................................................... 23
3.6.1.3.3 Design Loads for Decks, Deck Systems, and the Top Slab of Box Culverts ....................................... 24
3.6.1.3.4 Deck Overhang Load ............................................................................................................................ 24
3.6.1.4 FATIGUE LOAD ........................................................................................................................................... 25
3.6.1.4.2 Frequency ............................................................................................................................................. 25
3.6.1.5 RAIL TRANSIT LOAD ................................................................................................................................. 25
3.6.1.5.1P General ............................................................................................................................................... 25
3.6.1.5.2P Distribution of Rail Transit Loads Through Earth Fill ....................................................................... 25
3.6.1.6 PEDESTRIAN LOADS ................................................................................................................................. 29
3.6.2 Dynamic Load Allowance: IM .............................................................................................................................. 29
3.6.2.1 GENERAL ..................................................................................................................................................... 29
3.6.2.1.1P Components for which IM is Applicable ........................................................................................... 29
3.6.2.1.2P Components for which IM is Not Applicable .................................................................................... 29
3.6.2.2 BURIED COMPONENTS .............................................................................................................................. 30
3.6.2.3 WOOD COMPONENTS ................................................................................................................................. 30
3.6.4 Braking Force: BR................................................................................................................................................. 30
3.6.5 Vehicular Collision Force CT ................................................................................................................................. 31
3.6.5.3 Vehicular Collision with barriers .................................................................................................................... 31
3.8 WIND LOAD: WL AND WS .......................................................................................................................................... 31
3.8.1 Horizontal Wind Pressure...................................................................................................................................... 31
3.8.1.2 WIND PRESSURE ON STRUCTURES: WS ................................................................................................ 31
3.8.1.2.1 General .................................................................................................................................................. 31
C3.8.1.2.2 Loads from Superstructures ................................................................................................................ 32
3.8.3 Aeroelastic Instability ............................................................................................................................................. 32
3.8.3.4 WIND TUNNEL TESTS................................................................................................................................ 32
3.9 ICE LOADS: IC ............................................................................................................................................................... 32
3.9.1 General .................................................................................................................................................................... 32
3.9.5 Vertical Forces due to Ice Adhesion ....................................................................................................................... 32
3.10 EARTHQUAKE EFFECTS: EQ.................................................................................................................................. 33
C3.10.1 General................................................................................................................................................................ 33
3.10.2 Acceleration Coefficient ....................................................................................................................................... 34
3.10.7 Response Modification Factors............................................................................................................................ 35
3.10.7.1 GENERAL ................................................................................................................................................... 35
3.10.9 Calculation of Design Forces ............................................................................................................................... 35
3.10.9.2 SEISMIC ZONE 1 ........................................................................................................................................ 35
3.10.9.3 SEISMIC ZONE 2 ........................................................................................................................................ 36
3.10.9.5 LONGITUDINAL RESTRAINERS ............................................................................................................. 37
3.11 EARTH PRESSURE: EH, ES, LS AND DD ............................................................................................................... 37
3.11.1 General .................................................................................................................................................................. 37
3.11.3 Presence of Water ................................................................................................................................................. 37
3.11.5 Earth Pressure: EH ............................................................................................................................................. 37
C3.11.5.2 AT-REST PRESSURE COEFFICIENT, ko ............................................................................................... 37
C3.11.5.3 ACTIVE PRESSURE COEFFICIENT, ka ................................................................................................. 38
3.11.5.4 PASSIVE PRESSURE COEFFICIENT, kp.................................................................................................. 38
3.11.5.5 EQUIVALENT-FLUID METHOD OF ESTIMATING EARTH PRESSURES ......................................... 40
B.2 - cclxxxvi
3.11.5.6P EARTH PRESSURE FOR NONGRAVITY CANTILEVER WALLS ..................................................... 41
3.11.5.7 APPARENT EARTH PRESSURES FOR ANCHORED WALLS .............................................................. 47
3.11.5.6.1P Cohesionless Soils ............................................................................................................................ 48
3.11.5.7.2P Cohesive Soils .................................................................................................................................. 51
3.11.5.7.2a Stiff to Hard, Including Fissured Cohesive Soils ...................................................................... 51
3.11.5.6.2bP Very Soft to Medium-Stiff Cohesive Soils................................................................................ 52
3.11.6 Surcharge Loads: ES and LS .............................................................................................................................. 55
3.11.6.4 LIVE LOAD SURCHARGE: LS ................................................................................................................ 55
3.12 FORCE EFFECTS DUE TO SUPERIMPOSED DEFORMATIONS: TU, TG, SH, CR, SE ................................. 56
3.12.2 Uniform Temperature .......................................................................................................................................... 56
3.12.2.1.1 TEMPERATURE RANGES ..................................................................................................................... 56
3.12.7 Minimum Temperature Force for Fixed Substructures .................................................................................... 57
3.13 FRICTION FORCES: FR ............................................................................................................................................. 58
3.14 VESSEL COLLISION: CV .......................................................................................................................................... 58
3.14.1 General .................................................................................................................................................................. 58
3.14.2 Owner's Responsibility ......................................................................................................................................... 58
C3.14.15 Protection of Substructures ............................................................................................................................ 58
3.15P FORCE TRANSFER TO SUBSTRUCTURE ............................................................................................................ 59
3.15.1P Longitudinal Force ............................................................................................................................................. 59
3.15.1.2P FORCE TRANSFER TO SUBSTRUCTURE ........................................................................................... 59
3.15.1.3P EFFECTIVE LENGTH FOR SUPERSTRUCTURE FORCES ................................................................. 59
3.15.1.4P FORCE RESOLUTION TO SUBSTRUCTURE ....................................................................................... 59
3.15.2P Transverse Force ................................................................................................................................................ 60
3.15.2.2P EFFECTIVE LENGTHS FOR SUPERSTRUCTURE FORCES............................................................... 61
3.15.2.4P DETERMINATION OF BEARING REACTIONS ................................................................................... 61
B.2 - cclxxxvii
[THIS PAGE IS INTENTIONALLY LEFT BLANK]
B.2 - cclxxxviii
DM-4, Section 2 - General Design and Location Features
SPECIFICATIONS
September 2007
COMMENTARY
2.3 LOCATION FEATURES
2.3.2 Bridge Site Arrangement
2.3.2.2 TRAFFIC SAFETY
2.3.2.2.1 Protection of Structures
The following shall supplement A2.3.2.2.1.
The minimum lateral clearance for locating substructure
units for bridges over highways or railroads shall be as
follows:
(a) For bridges over a highway, refer to Design
Manual Part 2, Chapter 4.4.C for the minimum
horizontal clearance for locating substructure units.
(b) For bridges over railroads, see D2.3.3.4.
B.2 - 1
September 2007
SPECIFICATIONS
COMMENTARY
2.3.2.2.2 Protection of Users
C2.3.2.2.2
The following shall replace the last paragraph of
A2.3.2.2.2.
Sidewalks on bridges shall be protected by barriers
unless approved by the Department.
AC2.3.2.2.2.
An example where the Department may waive the
sidewalk barrier requirement would be in an urban
environment where a curbed approach walkways exists and
the posted vehicular speed is less than or equal to 50 km/hr
{30 mph}.
2.3.2.2.3 Geometric Standards
The following shall replace the last sentence of
A2.3.2.2.3.
For roadway geometry refer to Design Manual, Part 2.
For bridge-mounted barriers, structure-mounted guide rail
and other protective devices refer to appropriate Standard
Drawings.
2.3.3 Clearances
2.3.3.2 HIGHWAY VERTICAL
The following shall replace the first sentence of the first
paragraph of A2.3.3.2.
Refer to Design Manual, Part 2, for minimum vertical
clearance for overhead bridges.
The following shall supplement A2.3.3.2.
Vertical clearance over the width of the roadway,
including shoulders, shall be provided in accordance with
Design Manual, Part 2, Chapter 2, Section 2.21, and as
follows:
•
For bridges over railroads, see D2.3.3.4.
•
Minimum required vertical clearance shall preferably
be maintained within the recovery area. In calculating
actual vertical clearance under a beam splice, an
allowance of 20 mm {3/4 in.}, plus the thickness of the
outside flange splice plate, shall be made.
•
Vertical clearance is to be measured at any high point
on the outer edge shoulder. 30 mm {1 in.} will be
deducted from the actual measurement for posting
purposes to account for minor variations as proper
location of measurement and jumping of traveling
vehicles.
•
For vertical sag under the bridge, ensure that the
vertical clearance is measured from a chord between
any two high points along the traveling direction to
account for the maximum truck length permitted on the
road.
For prestressed concrete beams, do not take credit for
beam camber in determining actual vertical clearance,
B.2 - ii
SPECIFICATIONS
September 2007
COMMENTARY
unless the beam is cast or assembled specifically to provide
vertical curvature of the bottom of the beam.
2.3.3.3 HIGHWAY HORIZONTAL
The following shall supplement the first paragraph of
A2.3.3.3.
Refer to Design Manual, Part 2, for additional details on
bridge widths, including criteria for bridges on Very Low
Volume Roads.
2.3.3.4 RAILROAD OVERPASS
Pennsylvania Public Utility Commission (PUC) has
jurisdiction on railroad overpass clearances.
Refer to Design Manual, Part 1A, Chapter 7, Section
12, for additional details where railroad are overpassed by a
highway structure.
Structures carrying railroad tracks shall be designed
according to AREMA specifications and the modifications
adopted by the railroad system involved.
For structures carrying highways over railroad tracks,
the minimum horizontal clearance, specified and provided,
from the centerline of track shall be in accordance with
Publication 371 and/or Railroad Form D-4279 to face of
abutment or pier and shall be shown on the drawings. A
5500 mm {18'-0"} lateral clearance from the centerline of
track shall be provided for off-track equipment on one side
if requested by the railroad. Class 1 (major) railroads may
require additional lateral clearance depending upon the need
for drainage ditches and the roadway for off-track
equipment. If track and abutment or piers are skewed
relative to each other, horizontal clearances to the
extremities of the structure shall also be shown. If the track
is on a curve within 24 400 mm {80 ft.} of the crossing,
additional horizontal clearance is required to compensate for
the curve (refer to AREMA, Volume 2, Chapter 28). If a
railroad requests clearance in excess of the above, complete
justification of this request shall be provided.
The
agreement on the lateral and vertical clearances shall be
reached with the operating railroad, or the determination
from the PUC shall be secured prior to submitting for TS&L
approval.
The minimum vertical clearance over the top of rail
shall be in accordance with Publication 371 and/or Railroad
Form D-4279 and shall be shown for each track on the
drawings. If track and abutments or piers are skewed
relative to each other, vertical clearances to the extremities
of the structure should also be shown. Approval for any
exception to the above minimum clearance over railroad
tracks shall be secured from the operating railroad company
or the PUC prior to submitting for TS&L approval.
To provide for a drainage ditch parallel to track, the
elevation of the top of footings adjacent to track shall be at
least 1100 mm {3'-6"} below the elevation of the top of rail,
B.2 - iii
SPECIFICATIONS
September 2007
COMMENTARY
unless rock is encountered.
The edge of footing shall be at least 2200 mm {7 ft.}
from the centerline of adjacent track.
If pier bents are used between 5480 mm {18 ft.} and
7620 {25 ft.} from the centerline of tracks, columns shall be
protected by crash walls at least 760 mm {2’-6‖} thick,
which shall extend 3048 mm {10 ft.} above the top of rail
and 1828 mm {6 ft.} for single column or 762 mm {2’-6‖}
for multi-column bents beyond the outside face of outside
columns. The crash wall shall rest upon the column
footings, extend 152 mm {6 in.} from the face of columns
adjacent to traffic, and shall connect all columns in a pier
bent. Solid piers with a minimum thickness of 760 mm {2’6‖} and a length of 6100 mm {20 ft.}, single column piers
of minimum 1220 mm x 3800 mm {4’-0‖ x 12’-6‖}
dimensions or any solid pier sections with equivalent cross
sections and minimum 760 mm {2’-6‖} thickness negate the
need to provide crash walls. Reinforcement to be designed
in accordance with A3.6.5.2, but not less than horizontal
bars of No. 19 at 300 mm {No. 6 at 12 in.} each face, and
vertical stirrups of No. 13 at 300 mm {No. 4 at 12 in.}.
Crash walls meeting the same dimension and reinforcement
requirements as above shall also be provided in front of
prefabricated walls.
Bridge scuppers shall not drain onto railroad tracks.
Provision shall be made to direct surface water from the
bridge area into an adequate drainage facility along the
railroad track, in which case drainage approval by the
railroad company is required prior to submission of final
plans.
Safety provisions required during excavation in the
vicinity of railroad tracks and substructures shall be in
accordance with the Special Provision ―Maintenance and
Protection of Railroad Traffic‖.
Sheet piling used during excavation for protection of
railroad tracks and substructure shall be designed according
to AREMA specifications and shall be subject to approval
by the railroad company. The use of caisson footings shall
be evaluated in lieu of sheet piling and deep foundation.
Complete details of temporary track(s) or a temporary
railroad bridge to be constructed by the Department's
contractor shall be shown on the design drawings.
Applicable railroad design standards or design drawings
shall be referred to or duplicated on the design drawings.
For highway structures with sidewalks, protective
fencing shall be provided on all structures crossing over
railroads. The protective fence shall extend at least 2440
mm {8 ft.} from top of sidewalk or driving surface adjacent
to the barrier wall. The fence may be placed on top of the
barrier wall.
All railroad clearances shall be based on the railroad’s
current design criteria.
For electrified railroad tracks, these additional
requirements apply:
•
If a railroad is electrified, it shall be so noted on the
B.2 - iv
SPECIFICATIONS
September 2007
COMMENTARY
preliminary plans submitted for type, size and location
approval.
•
Protective barrier shall be provided on spans or on part
of spans for structures over electrified railroads, as
directed by the railroad company. Generally, the
protective barrier shall extend at least 3100 mm {10 ft.}
beyond the point at which any electrified railroad wire
passes under the bridge. However, in no case shall the
end of the protective barrier be less than 3100 mm {10
ft.} from the wire measured in a horizontal plane and
normal to the wire outside of the limit of the bridge, and
less than 1900 mm {6 ft.} from the wire within the limit
of the bridge.
•
Details of protective barriers are shown on Standard
Drawing BC-711M. If conditions warrant or if directed
by the railroad company, details shall be modified.
Such modifications shall be shown on the design
drawings.
•
All open or expansion joints in the concrete portion of
barriers, divisors, sidewalks, and curbs within the limits
of the barrier shall be covered or closed with joint
materials. Details of such joints shall be shown on the
design drawings.
•
In the case of bridges crossing electrified railroad
tracks, the details of catenary attachments and their
locations, if attached or pertinent to the structure, shall
be shown on the plans. Consideration shall be given to
realign the catenary by installing support columns on
each side of the bridge to avoid catenary attachments to
the bridge. Normally, ground cable attachments,
cables, miscellaneous materials, etc. are supplied by the
contractor and are installed by the railroad. A separate
block identifying materials required, description of
materials, railroad reference number for materials, and
party responsible for providing or installing materials
shall be shown on the plans. Approval of grounding
plans shall be obtained from the railroad concurrently
with approval of the structure drawings.
Where the PUC has jurisdiction over the structure
involved, PUC Docket Number, either A.____ or C.____ (A
stands for Application and C stands for complaint) shall be
shown on the first sheet of the design drawings (S-drawings)
above the title block, after the PUC has approved the plans.
The responsible designer shall add the PUC Docket
Number on all plans, and the District shall add the PUC
Docket Number in BMS where the PUC has jurisdiction
over the structure involved.
Where applicable, the USDOT/AAR number of the
existing structure shall be shown on the first sheet of the
design drawings.
2.5 DESIGN OBJECTIVES
B.2 - v
SPECIFICATIONS
September 2007
COMMENTARY
2.5.2 Serviceability
2.5.2.2 INSPECTABILITY
D2.5.2.2.1P, D2.5.2.2.2P, D2.5.2.2.3P and D2.5.2.2.4P
shall replace A2.5.2.2.
2.5.2.2.1P General
Inspection and maintenance instructions and
requirements for critical details shall be stipulated on the
plans. The plans shall also include reference to the method
of access for inspection of the subject details.
It is necessary to have adequate means of access for
bridge safety inspection.
Review bridge designs for
inspectability at TS&L, final design and construction stages.
PP3.6.6 provides information for checking a structures
inspectability with PennDOT's underbridge crane.
For special bridge conditions, the inspectability shall be
as determined by the Chief Bridge Engineer.
2.5.2.2.2P Inspection Walks
2.5.2.2.2aP General
Unless approved otherwise, inspection walks shall be
provided for long bridges (over 300 000 mm {1,000 ft.})
which cannot be readily inspected using inspection crane or
which are otherwise inaccessible from underneath.
Generally, inspection walks are required under the following
conditions:
•
Bridge width over 18 000 mm {60 ft.}, inaccessible
from underneath
•
Superstructure depth over 3500 mm {11.5 ft.},
including beams, barrier, railing or fencing, and noise
walls, inaccessible from underneath
•
High bridge underclearance (in excess of 9000 mm {30
ft.}), particularly for bridges over large rivers
B.2 - vi
SPECIFICATIONS
September 2007
COMMENTARY
2.5.2.2.2bP Design Live Load
A minimum design live load of 3.8 kN/m2 {80 psf}
shall be used.
2.5.2.2.2cP Geometry
The minimum width shall be 1200 mm {4 ft.}. The
minimum overhead clearance shall be 1800 mm {6 ft.}.
Toe guard protection and railing shall be provided if the
walk is not protected by the girders. The walk shall be
secured against vandalism and shall not provide entrance to
the general public. The entrances shall be locked or secured
against access. Provision shall be made to cross from one
bay to the next, generally at pier locations for at least one
person, 1800 mm {6 ft.} in height, carrying tools and
equipment.
2.5.2.2.2dP Connection to the Main Members
Generally, a bolted or threaded insert type of
connection shall be provided to secure the walks in position.
Lock washers or another positive connection device shall be
specified to protect the connection from being loosened due
to bridge vibration.
2.5.2.2.3P Inspectability for Enclosed Section
Vent holes and large size (600 mm by 900 mm {2 ft. by
3 ft.} minimum and 900 mm by 1200 mm {3 ft. by 4 ft.}
desirable opening) inspection hatches shall be provided for
large-span box structures; provision for lighting, cross
ventilation, and steps shall be made where required. Large
box sections shall have a coat of white paint on the interior.
2.5.2.2.4P Girder Handrail
Where other inspection facilities are not provided,
handrails shall be attached to the web of steel girders greater
than 1800 mm {72 in.} in depth.
2.5.2.2.5P Sound Barriers
Sound barriers shall be designed and detailed to
maintain bridge inspectability. For special conditions, the
inspectability shall be determined by the Chief Bridge
Engineer.
2.5.2.6 DEFORMATIONS
2.5.2.6.1 General
C2.5.2.6.1
The following shall replace A2.5.2.6.1.
Bridges shall be designed to avoid undesirable
structural or psychological effects due to their deformations.
The following shall replace AC2.5.2.6.1.
Service load deformations may cause deterioration of
wearing surfaces and local cracking in concrete slabs and in
metal bridges which could impair serviceability and
B.2 - vii
SPECIFICATIONS
September 2007
COMMENTARY
durability, even if self limiting and not a potential source of
collapse.
2.5.2.6.2 Criteria for Deflection
C2.5.2.6.2
In applying criteria for deflection, the vehicular load
shall include the dynamic load allowance.
To control deflections of structures, the following
principles shall apply:
•
when investigating the maximum absolute deflection,
all design lanes should be loaded, and all supporting
components should be assumed to deflect equally,
For a multi-beam bridge, this is equivalent to saying
that the distribution factor for deflection is equal to the
number of lanes divided by the number of beams.
•
for composite design, the design cross-section should
include the entire width of the roadway, neglecting any
stiffness contribution by barrier, railing or other
secondary members of the bridge,
The weight of barrier, railing or other secondary
members shall be included for deflection and design. Only
the stiffness of these items should be neglected.
•
when investigating maximum relative displacements,
the number and position of loaded lanes should be
selected to provide the worst differential effect,
•
the live load portion of Load Combination Service I of
Table A3.4.1-1 should be used, including the dynamic
load allowance, IM
•
the live load shall be taken from D3.6.1.3.2,
•
the provisions of A3.6.1.1.2 should apply,
•
for skewed bridges, a normal cross-section may be
used; for curved and curved skewed bridges a radial
cross-section may be used.
The following deflection limits shall be used for steel,
aluminum and/or concrete construction:
•
vehicular load, general ..................................... Span/800
•
vehicular and/or pedestrian loads .................. Span/1000
•
vehicular load on cantilever arms .................... Span/300
•
vehicular and/or pedestrian loads on
cantilever arms ................................................. Span/375
For steel I-shaped beams and girders, the provisions of
A6.10.4 and A6.11.4, respectively, regarding the control of
permanent deflections through flange stress controls, shall
apply.
The following deflection limits shall be used for wood
construction:
•
vehicular load, general ..................................... Span/425
From a structural viewpoint, large deflections in wood
components cause fasteners to loosen and brittle materials,
such as asphalt pavement, to crack and break. In addition,
members that sag below a level plane present a poor
B.2 - viii
SPECIFICATIONS
September 2007
COMMENTARY
•
vehicular and/or pedestrian loads .................... Span/800
•
vehicular load on wood planks and panels:
extreme relative deflection between
adjacent edges ...................................... 2.5 mm {1/8 in.}
The following provisions shall apply to orthotropic
plate decks:
•
vehicular load on deck plate ............................ Span/300
•
vehicular load on ribs of orthotropic metal
decks .............................................................. Span/1000
•
vehicular load on ribs of orthotropic
metal decks: extreme relative deflection
between adjacent ribs........................... 2.5 mm {1/8 in.}
appearance and can give the public a perception of structural
inadequacy. Deflections from moving vehicle loads also
produce vertical movement and vibrations that annoy
motorists and alarm pedestrians, Ritter (1990).
Excessive
deformation
can cause
premature
deterioration of the wearing surface and affect the
performance of fasteners, but limits on the latter have not
yet been established.
The intent of the relative deflection criterion is to
protect the wearing surface from debonding and fracturing
due to excessive flexing of the deck. The restriction on
relative rib displacement may be revised or removed when
more data is available to formulate appropriate requirements
as function of thickness and physical properties of the
wearing surface employed.
2.5.2.6.3 Criteria for Span-to-Depth Ratios
The following shall replace the first paragraph of
A2.5.2.6.3.
Structures or components of structures shall satisfy the
span-to-depth ratios given in Table A2.5.2.6.3-1
where:
S
=
slab span length (mm) {ft}
L
=
span length (mm) {ft}
Concrete decks on multi-girder-type bridges shall satisfy the
span-to-depth ratios in Table A1 with the heading "Slabs".
2.5.2.7 CONSIDERATION OF FUTURE WIDENING
2.5.2.7.1 Exterior Beams on Multi-Beam Bridges
C2.5.2.7.1
The load carrying capacity of exterior beams shall not
be less than the load carrying capacity of an interior beam,
unless specifically approved by the Chief Bridge Engineer.
The stiffness of the interior and exterior beams should
be relatively equal.
2.5.3 Constructibility
C2.5.3
The following shall supplement A2.5.3.
An acceptable slab placement sequence shall be shown
in the contract plans. Figure 1 illustrates the format to be
used. The actual sequence shall be determined from an
erection analysis (see D6.10.3.2.4P) for the specific
structure in question. For steel girder structures and
prestressed beams made continuous for live load, see
D6.10.3.2.4P and D5.14.1.2.7fP, respectively, for additional
requirements concerning slab placement sequence. The
The two-curing day waiting period between pours in
adjacent continuous spans is for crack control. Studies have
shown that longer waiting periods have no significant effect
on cracking, primarily because shrinkage is the dominating
factor in cracking. The two-curing day period between
adjacent pours in the same span will provide enough
strength gain to introduce composite action and will increase
the stability of the girder over the length of the previous
B.2 - ix
September 2007
SPECIFICATIONS
COMMENTARY
required curing strength (if any) of the concrete of a
previous placement segment shall be designated as
appropriate. Instead of specifying a curing strength of the
concrete, a time delay (if any) between placements may be
designated, as appropriate. A minimum waiting period of
two curing days between positive moment region
placements in immediately adjacent continuous spans and
between adjacent positive moment placements in the same
span shall be specified in the contract plans. A minimum
waiting period between other placements need not be
specified if analysis suggests that it is unnecessary.
pour.
Figure 2.5.3-1 - Example of Slab Placement Sequence for Contract Plans
The Contractor may use an alternate slab placement
sequence if the following provisions are met:
The Contractor shall submit to the Department a revised
slab placement sequence with support calculations and
computer stress analysis. The revised slab placement
sequence shall meet the requirements of which the
original slab placement sequence were based on.
The Department will review and approve calculations.
The Contractor shall receive written approval prior to
the use of the revised slab placement sequence and/or
camber values.
All costs for the development and approval of the
revised slab placement sequence and camber values
shall be borne by the Contractor.
The Department will be the sole judge of the
acceptability of the revised slab placement sequence
and camber values.
2.5.3.1P FALSEWORK
C2.5.3.1P
B.2 - x
SPECIFICATIONS
September 2007
COMMENTARY
Composite beams shall be designed with no
intermediate falsework during placing and curing of the
concrete deck.
When falsework is used on a project, it shall be
designed for the following items, but not limited to:
For overhang bracket requirements see D6.10.3.2.4.2P.
vertical loads,
horizontal loads,
differential settlement forces,
unbalanced temporary
construction), and
loadings
(e.g.,
staged
errant highway vehicles.
The following guidelines should be used for the
approval of the falsework:
Refer to Appendix P for guidance on jacking and
supporting the superstructure.
Every bridge on a project should receive a separate
falsework design analysis.
In the event falsework is moved from one bridge to
another, it should be thoroughly inspected for structural
damage and plumbness to ensure that all members are
in place and properly aligned and corrected.
Ensure that the requirement of Publication 408, Section
105.02(c), ―all drawings for load bearing falsework
submissions are to be signed and sealed by a
Professional Engineer, registered in the Commonwealth
of Pennsylvania‖, is fully enforced.
During the falsework review, make sure that it is
designed to handle vertical and horizontal loading and
to contain enough redundancy to prevent a failure in the
entire system.
Vertical loading and differential
settlement forces, and horizontal lateral and
longitudinal forces should be taken into account.
Unbalanced temporary loading caused by the placement
sequence, should also be considered.
If an unfortunate event occurs due to the failure of the
falsework, preserve and document the in-place failure
and assign investigation responsibilities to qualified
impartial parties.
If service load design is used, designers may increase
the allowable basic unit stress by 25% for temporary
falsework.
For purposes of these guidelines, temporary falsework
is defined as falsework constructed for no more than one
construction season.
Should the contractor deem that temporary falsework is
necessary for the construction of curved and skewed steel
While using no temporary falsework is desirable from a
cost-effectiveness perspective, should the designer and/or
B.2 - xi
September 2007
SPECIFICATIONS
COMMENTARY
bridges, the following guidelines should be used for its
placement:
contractor deem that falsework is needed to ensure that a
curved or skewed steel bridge is constructible, it should
initially be placed near splice locations. When girder
vertical deflections are still a concern, an additional
temporary support should be placed as close as possible to
the location of maximum vertical deflection of the span
(approximately 0.4 L from an abutment, where L is the span
length, for side spans and 0.5 L for intermediate spans) to
reduce girder deflections. For the side spans, when adding
multiple temporary supports is not feasible, placing one
support near 0.75L from the abutments is suggested.
When temporary falsework is needed for a span, it shall
be placed at locations to reduce splice rotations and
girder vertical deflections.
The stability of the structure supported by temporary
falsework shall be evaluated.
C2.5.3.2P
2.5.3.2P GIRDER ERECTION SEQUENCE
Paired girder erection, as opposed to single erection,
requires fewer temporary supports for the erected segments
during all stages of construction. However for bridges with
an odd number of girders, at least one girder line must be
erected by itself.
The following guidelines should be used for the girder
erection sequence for horizontally curved steel I-girder
structures:
Should adequate crane capacity be available, paired
girder erection approaches are preferred.
When the radius of the curved structure is less than 300
feet, it is recommended that girders be placed from
inner radius to outer radius.
structure is stable for all stages of construction and that
supports necessary to maintain stability have been
provided.
The following guidelines should be used for the girder
erection sequence for skewed steel I-girder structures:
structure is stable for all stages of construction and that
supports necessary to maintain stability have been
provided.
Girder erection from inner radius to outer radius, when
compared to the opposite direction, can result in slightly
smaller deformations for the girders for all stages of
construction which, in turn, means the structure is more
constructible. This effect is more pronounced in severely
curved structures (i.e., radius less than 300 ft).
Stability, which refers to the prevention of excessive
deformations and the possibility of buckling, of partial and
completed girders at various stages of erection, is the
responsibility of the contractor as specified in Publication
408 Section 1050.3(c).
For construction of straight skewed bridges, paired erection
does require a smaller number of temporary supports but
offers no other substantial benefits over a single erection
approach with respect to deformations.
2.6 HYDROLOGY AND HYDRAULIC
2.6.6 Roadway Drainage
2.6.6.1 GENERAL
Dimensions for the deck cross slopes shall be shown in
the same manner as indicated on the roadway plans (e.g.,
2%). The water table cross slope on bridge decks which are
not superelevated shall be sloped toward the curb or median.
The rate of slope shall be 4% for water table widths of
1800 mm {6 ft.} or less, and 3% for water table widths over
1800 mm {6 ft.}. On superelevated decks, the water table
on the high side shall be as specified in Design Manual,
Part 2. On the low side, the water table shall slope in the
B.2 - xii
SPECIFICATIONS
September 2007
COMMENTARY
same direction and magnitude as the adjacent lane, but not
less than 4% for water table widths of 1800 mm {6 ft.} or
less, nor less than 3% for water table widths over 1800 mm
{6 ft.}. On a superelevated bridge with a paved median,
adjustment of the grades of adjacent roadways may be
required to equalize the height of the divisor or median
barrier.
2.7P DESIGN DRAWINGS
2.7.1P Moment and Shear Envelope Diagrams
For simple span bridges, the contract plans shall contain
a table of following items:
(1) Maximum non-composite dead load moment
(2) Maximum composite dead load moment (including
future wearing surface)
(3) Maximum live load plus impact moment for PHL93 and P-82 loading conditions
(4) Maximum non-composite dead load shear
(5) Maximum composite dead load shear (including
(6) Maximum live load plus impact shear for PHL-93
and P-82 loading conditions
(7) Composite and non-composite section properties at
resisting sections
For multiple span continuous bridges, as a minimum,
the contract plans shall contain a diagrammatic presentation
of the following on a per-girder basis:
(1) Non-composite dead load moment diagram
(2) Composite dead load moment diagram (including
(3) Separate positive and negative live load plus
impact moment envelope for PHL-93 and P-82
loading conditions
(4) Summation of (1) and (2)
(5) Non-composite dead load shear diagram
(6) Composite dead load shear diagram (including
(7) Separate positive and negative live load plus
impact shear envelope for PHL-93 and P-82
B.2 - xiii
SPECIFICATIONS
September 2007
COMMENTARY
loading conditions
(8) Summation of (5) and (6)
(9) Composite and non-composite section properties at
the resisting sections
The data to construct this presentation shall not have
load factors applied.
Also, a table of reactions shall be provided for total
dead load, and positive and negative live load, plus impact
without load factors applied.
2.7.2P Major, Complex and Unusual Bridges
2.7.2.1P LOAD DATA SHEET
For the new design of major, complex and unusual
bridges, additional information shall be included on the
design drawings for typical common components such as
bearings, floorbeams, and stringers. For these items as a
minimum, the contractor plans shall contain a tabular
presentation of the following:
Bearings
(a) Vertical Force
1.
2.
3.
Total Dead Load
Live Load plus Impact
Summation of 1. and 2.
(b) Transverse Force Wind Load
(c) Longitudinal Force
1.
2.
3.
Wind Load
Traction Load
Friction Load
Truss Members
(a) Axial Force
1.
2.
3.
Total Dead Load
(b) Bending Moment
1.
2.
3.
Total Dead Load
(c) Section Properties
B.2 - xiv
SPECIFICATIONS
1.
2.
3.
September 2007
COMMENTARY
Gross Area
Net Area
Section Modulus
Floorbeam
Provide the same type of information as required in
D2.7.1P for girders.
Stringers
Provide the same type of information as required in
D2.7.1P for girders.
The data for this tabular presentation shall not have load
factors applied.
2.8P BRIDGE SECURITY
2.8.1P General
For the purpose of this section bridges deemed
important shall include (1) new singular bridges of total
replacement value exceeding $100 million or (2) new or
existing bridges identified as critical by the Department’s
Emergency Transportation Operations Section.
A risk management approach shall be utilized to assess
structural vulnerability and countermeasures. A threat based
component level analysis shall be conducted considering a
full range of threats. FHWA’s workshop methodology or
other BQAD approved methodology shall be used.
The results of such processes and included
design/mitigation criteria and features shall be considered
sensitive project information that is protected by restricted
access. Contract documents shall not include reference to
any security standard, design capability, or other
information that might provide knowledge of bridge
resistance.
For all bridges, restrict access to doors/hatches by using
locks or by making inaccessible accept by special mobile
equipment, such as snooper, man lift, etc. In cellular
structures, avoid vent hole diameters larger than 2 inches
when holes are in easy reach. Avoid nooks and areas that
allow for concealed access and create a confined pressure
effect. If these details are unavoidable consideration should
be given to barring access.
B.2 - xv
SPECIFICATIONS
September 2007
COMMENTARY
PENNSYLVANIA DEPARTMENT OF TRANSPORTATION
DESIGN MANUAL
PART 4
VOLUME 1
SECTION 3 – LOADS AND LOAD FACTORS
2.3 LOCATION FEATURES .................................................................................................................................................. 1
2.3.2 Bridge Site Arrangement ......................................................................................................................................... 1
2.3.2.2 TRAFFIC SAFETY .......................................................................................................................................... 1
2.3.2.2.1 Protection of Structures .......................................................................................................................... 1
2.3.2.2.2 Protection of Users ................................................................................................................................. ii
2.3.2.2.3 Geometric Standards .............................................................................................................................. ii
2.3.3 Clearances ................................................................................................................................................................. ii
2.3.3.2 HIGHWAY VERTICAL .................................................................................................................................. ii
2.3.3.3 HIGHWAY HORIZONTAL ........................................................................................................................... iii
2.3.3.4 RAILROAD OVERPASS ............................................................................................................................... iii
2.5 DESIGN OBJECTIVES ..................................................................................................................................................... v
2.5.2 Serviceability ............................................................................................................................................................ vi
2.5.2.2 INSPECTABILITY ......................................................................................................................................... vi
2.5.2.2.1P General ................................................................................................................................................ vi
2.5.2.2.2P Inspection Walks ................................................................................................................................. vi
2.5.2.2.2aP General ...................................................................................................................................... vi
2.5.2.2.2bP Design Live Load ...................................................................................................................... vii
2.5.2.2.2cP Geometry ................................................................................................................................... vii
2.5.2.2.2dP Connection to the Main Members ............................................................................................. vii
2.5.2.2.3P Inspectability for Enclosed Section .................................................................................................... vii
2.5.2.2.4P Girder Handrail .................................................................................................................................. vii
2.5.2.2.5P Sound Barriers .................................................................................................................................... vii
2.5.2.6 DEFORMATIONS ......................................................................................................................................... vii
2.5.2.6.1 General ................................................................................................................................................. vii
2.5.2.6.2 Criteria for Deflection ......................................................................................................................... viii
2.5.2.6.3 Criteria for Span-to-Depth Ratios.......................................................................................................... ix
2.5.2.7 CONSIDERATION OF FUTURE WIDENING ............................................................................................. ix
2.5.2.7.1 Exterior Beams on Multi-Beam Bridges ............................................................................................... ix
2.5.3 Constructibility ........................................................................................................................................................ ix
2.5.3.1P FALSEWORK ............................................................................................................................................... x
2.6 HYDROLOGY AND HYDRAULIC ............................................................................................................................... xii
2.6.6 Roadway Drainage.................................................................................................................................................. xii
2.6.6.1 GENERAL ..................................................................................................................................................... xii
2.7P DESIGN DRAWINGS .................................................................................................................................................. xiii
2.7.1P Moment and Shear Envelope Diagrams ............................................................................................................ xiii
2.7.2P Major, Complex and Unusual Bridges .............................................................................................................. xiv
2.7.2.1P LOAD DATA SHEET ................................................................................................................................ xiv
2.8P BRIDGE SECURITY ..................................................................................................................................................... xv
2.8.1P General .................................................................................................................................................................. xv
C3.1 SCOPE .............................................................................................................................................................................. 1
3.3 NOTATION......................................................................................................................................................................... 1
3.3.1 General ...................................................................................................................................................................... 1
3.4 LOAD FACTORS AND COMBINATIONS .................................................................................................................... 1
B.3 - xvi
SPECIFICATIONS
September 2007
COMMENTARY
3.4.1 Load Factors and Load Combinations ................................................................................................................... 1
3.4.1.1P LOAD FACTORS AND COMBINATIONS FOR TYPICAL PENNDOT BRIDGE COMPONENTS ....... 5
3.5 PERMANENT LOADS .................................................................................................................................................... 19
3.5.1 Dead Loads: DC, DW and EV .............................................................................................................................. 19
3.5.1.1P APPLICATION OF DEAD LOAD ON GIRDER AND BOX BEAM STRUCTURES .............................. 19
3.6 LIVE LOADS .................................................................................................................................................................... 19
3.6.1 Gravity Loads: LL, PL .......................................................................................................................................... 19
3.6.1.1 VEHICULAR LIVE LOAD ........................................................................................................................... 19
3.6.1.1.2 Multiple Presence of Live Load ............................................................................................................ 19
3.6.1.2 DESIGN VEHICULAR LIVE LOAD ........................................................................................................... 20
3.6.1.2.1 General ................................................................................................................................................. 20
3.6.1.2.3 Design Tandem..................................................................................................................................... 20
3.6.1.2.5 Tire Contact Area ................................................................................................................................. 20
3.6.1.2.6 Distribution of Wheel Loads through Earth Fills ................................................................................. 21
3.6.1.2.7P Design Permit Load ............................................................................................................................ 21
3.6.1.2.8P Maximum Legal Load (ML-80) ......................................................................................................... 22
3.6.1.3 APPLICATION OF DESIGN VEHICULAR LIVE LOADS ........................................................................ 22
3.6.1.3.1 General ................................................................................................................................................. 22
3.6.1.3.2 Loading for Live Load Deflection Evaluation ...................................................................................... 23
3.6.1.3.3 Design Loads for Decks, Deck Systems, and the Top Slab of Box Culverts ....................................... 24
3.6.1.3.4 Deck Overhang Load ............................................................................................................................ 24
3.6.1.4 FATIGUE LOAD ........................................................................................................................................... 25
3.6.1.4.2 Frequency ............................................................................................................................................. 25
3.6.1.5 RAIL TRANSIT LOAD ................................................................................................................................. 25
3.6.1.5.1P General ............................................................................................................................................... 25
3.6.1.5.2P Distribution of Rail Transit Loads Through Earth Fill ....................................................................... 25
3.6.1.6 PEDESTRIAN LOADS ................................................................................................................................. 29
3.6.2 Dynamic Load Allowance: IM .............................................................................................................................. 29
3.6.2.1 GENERAL ..................................................................................................................................................... 29
3.6.2.1.1P Components for which IM is Applicable ........................................................................................... 29
3.6.2.1.2P Components for which IM is Not Applicable .................................................................................... 29
3.6.2.2 BURIED COMPONENTS .............................................................................................................................. 30
3.6.2.3 WOOD COMPONENTS ................................................................................................................................. 30
3.6.4 Braking Force: BR................................................................................................................................................. 30
3.6.5 Vehicular Collision Force CT ................................................................................................................................. 31
3.6.5.3 Vehicular Collision with barriers .................................................................................................................... 31
3.8 WIND LOAD: WL AND WS .......................................................................................................................................... 31
3.8.1 Horizontal Wind Pressure...................................................................................................................................... 31
3.8.1.2 WIND PRESSURE ON STRUCTURES: WS ................................................................................................ 31
3.8.1.2.1 General .................................................................................................................................................. 31
C3.8.1.2.2 Loads from Superstructures ................................................................................................................ 32
3.8.3 Aeroelastic Instability ............................................................................................................................................. 32
3.8.3.4 WIND TUNNEL TESTS................................................................................................................................ 32
3.9 ICE LOADS: IC ............................................................................................................................................................... 32
3.9.1 General .................................................................................................................................................................... 32
3.9.5 Vertical Forces due to Ice Adhesion ....................................................................................................................... 32
3.10 EARTHQUAKE EFFECTS: EQ .................................................................................................................................. 33
C3.10.1 General................................................................................................................................................................ 33
3.10.2 Acceleration Coefficient ....................................................................................................................................... 34
3.10.7 Response Modification Factors............................................................................................................................ 35
3.10.7.1 GENERAL ................................................................................................................................................... 35
3.10.9 Calculation of Design Forces ............................................................................................................................... 35
3.10.9.2 SEISMIC ZONE 1 ........................................................................................................................................ 35
3.10.9.3 SEISMIC ZONE 2 ........................................................................................................................................ 36
3.10.9.5 LONGITUDINAL RESTRAINERS ............................................................................................................. 37
3.11 EARTH PRESSURE: EH, ES, LS AND DD ............................................................................................................... 37
B.3 - xvii
SPECIFICATIONS
September 2007
COMMENTARY
3.11.1 General .................................................................................................................................................................. 37
3.11.3 Presence of Water ................................................................................................................................................. 37
3.11.5 Earth Pressure: EH ............................................................................................................................................. 37
C3.11.5.2 AT-REST PRESSURE COEFFICIENT, ko ............................................................................................... 37
C3.11.5.3 ACTIVE PRESSURE COEFFICIENT, ka ................................................................................................. 38
3.11.5.4 PASSIVE PRESSURE COEFFICIENT, kp.................................................................................................. 38
3.11.5.5 EQUIVALENT-FLUID METHOD OF ESTIMATING EARTH PRESSURES ......................................... 40
3.11.5.6P EARTH PRESSURE FOR NONGRAVITY CANTILEVER WALLS ..................................................... 41
3.11.5.7 APPARENT EARTH PRESSURES FOR ANCHORED WALLS .............................................................. 47
3.11.5.6.1P Cohesionless Soils ............................................................................................................................ 48
3.11.5.7.2P Cohesive Soils .................................................................................................................................. 51
3.11.5.7.2a Stiff to Hard, Including Fissured Cohesive Soils ...................................................................... 51
3.11.5.6.2bP Very Soft to Medium-Stiff Cohesive Soils................................................................................ 52
3.11.6 Surcharge Loads: ES and LS .............................................................................................................................. 55
3.11.6.4 LIVE LOAD SURCHARGE: LS ................................................................................................................ 55
3.12 FORCE EFFECTS DUE TO SUPERIMPOSED DEFORMATIONS: TU, TG, SH, CR, SE ................................. 56
3.12.2 Uniform Temperature .......................................................................................................................................... 56
3.12.2.1.1 TEMPERATURE RANGES ..................................................................................................................... 56
3.12.7 Minimum Temperature Force for Fixed Substructures.................................................................................... 57
3.13 FRICTION FORCES: FR ............................................................................................................................................. 58
3.14 VESSEL COLLISION: CV .......................................................................................................................................... 58
3.14.1 General .................................................................................................................................................................. 58
3.14.2 Owner's Responsibility ......................................................................................................................................... 58
C3.14.15 Protection of Substructures ............................................................................................................................ 58
3.15P FORCE TRANSFER TO SUBSTRUCTURE ............................................................................................................ 59
3.15.1P Longitudinal Force ............................................................................................................................................. 59
3.15.1.3P EFFECTIVE LENGTH FOR SUPERSTRUCTURE FORCES ................................................................. 59
3.15.2P Transverse Force ................................................................................................................................................ 60
3.15.2.2P EFFECTIVE LENGTHS FOR SUPERSTRUCTURE FORCES............................................................... 61
3.15.2.4P DETERMINATION OF BEARING REACTIONS ................................................................................... 61
B.3 - xviii
SPECIFICATIONS
September 2007
COMMENTARY
[THIS PAGE IS INTENTIONALLY LEFT BLANK]
B.3 - xix
DM-4, Section 3 – Loads and Load Factors
SPECIFICATIONS
September 2007
COMMENTARY
C3.1 SCOPE
The following shall supplement the second paragraph of
AC3.1.
Before any test results are used in the design of a
structure, the tests and the test results must be approved by
the Chief Bridge Engineer
3.3 NOTATION
3.3.1 General
The following shall supplement A3.3.1
B
C
D
=
=
=
F
=
If =
kp =
=
M =
M =
Pa =
Pa =
Po =
Pp =
Pp =
Sm =
Su =
Wl =
Β =
β’ =
Vertical element width (mm) {ft} (3.11.5.6)
Pressure coefficient for loads applied on a subgrade (dim) (3.6.1.5.2P)
Depth of embedment of discrete and continuous vertical wall elements (mm) {ft} (3.11.5.6); width of uniformly
loaded area (mm) {ft} (3.6.1.5.2P)
Fictitious force applied at bottom of embedded continuous vertical wall element to provide horizontal force
equilibrium for simplified earth pressure distributions (N/mm) {kip/ft} (3.11.5.6)
Impact factor (dim) (3.6.1.5.2P)
Passive coefficient of lateral earth pressure (dim) (3.11.5.6)
Spacing between vertical wall elements (c/c) (mm) {ft} (3.11.5.6)
Length of uniformly loaded area (mm) {ft} (3.6.1.5.2P)
constant used in calculating active earth pressure coefficient in certain conditions (dim) (3.11.5.6.2bP)
Active resistance per vertical wall element (N) {kips}
Active earth pressure per unit length of wall (N/mm) {kip/ft} (3.11.5.6)
Intensity of the distributed load at the bottom of the railroad ties (MPa) {ksi} (3.6.1.5.2P)
Passive resistance per vertical wall element (N) {kips}
Passive earth pressure per unit length of wall (N/mm) {kip/ft} (3.11.5.6)
Shear strength of rock mass (MPa) {ksf} (3.11.5.6)
Undrained shear strength of cohesive soil (MPa) {ksf} (3.11.5.6)
Live load on structure from railroad loading (N/mm) {kip/ft} (3.6.1.5.2P)
Ground surface slope behind wall {+ for slope up from wall; - for slope down from wall} (DEG) (3.11.5.6)
Ground surface slope in front of wall {+ for slope up from wall; - for slope down from wall} (DEG) (3.11.5.6)
3.4 LOAD FACTORS AND COMBINATIONS
3.4.1 Load Factors and Load Combinations
C3.4.1
The following shall replace Strength II description in
A3.4.1.
The following shall replace Strength II commentary in
AC3.4.1.
Strength II - Load combination relating to the Design
Permit Load (P-82) use of the bridge.
This load combination only applies to the
superstructure, except for pier caps which
support a superstructure with a span
length greater than 20 000 mm {65 ft.}.
Bearings, (including uplift check),
substructure and foundation need not be
designed for this load combination. For
In design, the use of distribution factors in D4.6.2.2 and
A4.6.2.2 represents that the P-82 is in all design lanes.
The method for rating takes into account that the P-82 is
in one lane and the other lanes are occupied by the vehicular
live load.
In AC3.4.1, the commentary for Strength II states that
"For bridges longer than the permit vehicle, the presence of
the design lane load, preceding and following the permit
load in its lane, should be considered." A study done for the
B.3 - 1
SPECIFICATIONS
COMMENTARY
design, the distribution factors given in
D4.6.2.2 and A4.6.2.2 shall be used.
For the rating of existing bridges with Strength II
criteria, the following equation may be used to determine
Strength II live load moments and shear:
FRT
FRP
82
g1
1.2
FRPHL
September 2007
93
g
g1
1.2
Department showed that the P-82 with the interrupted lane
load only controls for moments in a small range of spans
and is only maximum of 2% above the PHL-93 loading. For
shear, the maximum difference between the PHL-93 and P82 with lane load was 7.5% with P-82 and lane load being
greater than PHL-93. The Department concluded that this
difference was acceptable because the study considered all
the lanes loaded with the P-82. Therefore, the P-82 loading
need not be considered with a partial lane load.
where:
FRT
=
total force response, moment or shear
FRP-82
=
P-82 force response, moment or shear
FRPHL-93 =
PHL-93 force response, moment or shear
g1
=
single lane distribution factor, moment or
shear
g
=
multi-lane distribution factor, moment or shear
FRT need not be taken greater than FRP-82(g).
The following shall supplement AC3.4.1.
STRENGTH IP -
Load combination relating to
the pedestrian load and a
reduced vehicular live load.
STRENGTH VI -
Load combination relating to
the design of piers which
includes ice and wind load
acting together.
EXTREME EVENT III - Load combination relating to
the failure of one element of
a component without the
failure of the component.
Extreme Events III and IV are uncalibrated load
combinations. They are intended to force consideration of
the safety of damaged structures.
EXTREME EVENT IV - Load combination relating to
the failure of one component
without the collapse of the
structure.
For this extreme event, a 3-D analysis is required. The
objective of this analysis is survival of the bridge (i.e., the
bridge may have large permanent deflections, but it has not
collapsed).
The conditions for which Extreme Event III and IV are
to be investigated are given in D1.3.4.
B.3 - 2
SPECIFICATIONS
September 2007
COMMENTARY
Table 3.4.1P-1 - Additional PennDOT Load Combinations and Load Factors
DC
DD
DW
EH
EV
ES
LL
IM
CE
BR
PL
LS
WA
STRENGTH IP
γp
*
-
-
-
STRENGTH VI
γp
-
1.25
1.25
EXTREME EVENT III
γp
γ'LL
-
EXTREME EVENT IV
γ' p
γ'LL
-
Load Combination
Limit State
WS
WL
FR
TU
CR
SH
TG
-
-
-
-
-
-
-
-
-
-
-
-
SE Use One of These at a Time
EQ
IC
CT
CV
-
-
-
-
-
-
-
-
1.25
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
*γLL = 1.35, γPL = 1.75
Table 3.4.1P-2 - Load Factor for Live Load for Extreme III and IV,
γ'LL
III
IV
γ'LL
γ'LL
PHL-93 Loading – all applicable
lanes
1.30
1.15
Permit load in governing lane with
PHL-93 in other applicable lanes
1.10
1.05
Case
Table 3.4.1P-3 - Load Factors for Permanent Loads for Extreme
Event IV, γ' p
Load Factor
Type of Load
Maximum
Minimum
DC: Component and Attachments
1.05
0.95
DW: Wearing Surfaces and Utilities
1.05
0.90
Unless otherwise specified, interaction of force effects
shall be accounted for by selecting load factors which
maximize and minimize each of the force effects one at a
time with the same load factors used to compute the
associated force effect.
As an example for a design which involves the
interaction of moment and axial force, the following four
design cases would be investigated:
select the load factors which maximize moment and use
these load factors in determining axial force
select the load factors which minimize moment and use
these load factors in determining axial force
select the load factors which maximize axial force and
use these load factors in determining moment
B.3 - 3
SPECIFICATIONS
September 2007
COMMENTARY
select the load factors which minimize axial force and
use these load factors in determining moment.
Due to the nature of force interaction, the absolute
worst case may not necessarily be that for which one of the
force effects is maximized, but an intermediate case.
However, the difference between the absolute worst case
and the design cases presented here are believed to be within
the tolerance of the design process. Therefore, as a
reasonable interpretation of the specification, maximum and
minimum force effects taken in conjunction with associated
force effects for interaction are to be considered. If the
Engineer believes that an intermediate case will govern to
an appreciable degree, the Engineer shall notify the Chief
Bridge Engineer. Then, the Chief Bridge Engineer will
determine if intermediate cases shall be investigated.
For MSE wall designs, D11.10.5.2 and D11.10.6.2 state
when to apply maximum and minimum EH and EV.
The following shall supplement the sixth paragraph of
A3.4.1 relating to TU, CR and SH.
The larger load factor shall be used to determine the
final length of the member. The smaller load factor shall be
used in determining force effects, such as creep and
shrinkage effects in pier caps and columns.
The following shall replace the ninth paragraph of
A3.4.1 relating to γTG and γSE.
For the application of temperature gradient see D3.12.3.
The load factor for settlement γSE shall be determined on a
project-specific basis.
The following shall replace the eleventh paragraph of
A3.4.1 relating to γEQ.
The load factor γEQ for live load for the Extreme EventI limit state shall be taken as 0.0.
The Department is currently using γEQ = 0.0 in
accordance with numerous past years of AASHTO practice.
We will continue to use γEQ = 0.0 until further work is
completed justifying a different value.
The following shall supplement A3.4.1 for the design of
box culverts.
Lateral earth pressures for box culverts shall be
computed using the equivalent fluid method given in
A3.11.5.5 and D3.11.5.5, and appropriate load factors, EH,
as given in Table 3.4.1-2, for horizontal earth pressures.
To maximize the load effect, the maximum at-rest load
factor shall be used with the maximum equivalent fluid
weight from Table D3.11.5.5-2, and the minimum at-rest
load factor shall be used with the minimum equivalent fluid
weight. In addition, a 50% reduction in both maximum and
minimum unfactored lateral earth pressures, EH and ES,
shall be considered for determining the maximum positive
moment in the top slab of the culvert, as specified in
A3.11.7.
In Table A3.4.1-2, the first bulleted item under EV:
Vertical Earth Pressure, ―Retaining Structures‖, shall be
changed to ―Retaining Walls and Abutments‖.
The following shall supplement AC3.4.1 for the design
of box culverts.
Rigid frame action of box culvert structures is assumed
to result in relatively small movements as compared to that
of a retaining wall or abutment-type structure, thus, an atrest condition is assumed.
B.3 - 4
SPECIFICATIONS
September 2007
COMMENTARY
3.4.1.1P LOAD FACTORS AND COMBINATIONS FOR
TYPICAL PENNDOT BRIDGE COMPONENTS
Tables 1 through 6 provide the load factors with the
corresponding limit state condition for the following typical
PennDOT bridge components:
steel girders (Table 1)
prestressed girders (Table 2)
abutment/retaining walls (Table 3)
box culverts (Table 4)
steel floorbeams (Table 5)
steel trusses (Table 6)
Tables 1, 2, 4, 5 and 6 also include information for
rating these components. (Rating are not typically done for
abutment/retaining walls.)
B.3 - 5
C3.4.1.1P
The design live load vehicle in the fatigue load
combination designated as HS20-9.0 refers to an HS20
truck with a fixed 9 meter {30 ft.} rear axle spacing.
DM-4 Section 3 – Loads and Load Factors
September 2007
Table D3.4.1.1P-1 - Load Factors and Live Load Vehicles for Steel Girders
Load Combination
STR I
STR
IP8
STR
IA6
STR
II
STR
III
STR
IV1
STR V
SERV
II
SERV
IIA6
SER
V IIB
FATIGUE2
DEFL2
CONST/
UNCURED
SLAB9
γDC3
1.25,
0.90
1.25,
0.90
1.25,
0.90
1.25,
0.90
1.25,
0.90
1.5
1.25,
0.90
1.00
1.00
1.00
---
---
1.25
γDW4
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.00
1.00
1.00
---
---
1.50, 0.65
γLL
1.75
1.35
1.10
1.35
---
---
1.35
1.30
1.00
1.00
0.75
1.00
1.50
γPL
---
1.75
---
---
---
---
---
---
---
---
---
---
---
γWS
---
---
---
---
1.40
---
0.40
---
---
---
---
---
1.25
PHL-93
PHL93
PHL-93
Permi
t
(P-82)
---
---
PHL93
PHL93
PHL93
Permi
t
(P82)
HS20-9.0
PennDOT
Defl. Trk.
User Def.
Design LL
Veh.7
Rating Veh.
PHL-93
Rating Applicability: I = Inventory, O = Operating
I
I
O
---
---
---
---
I
O
---
---
---
---
---
---
---
O
---
---
---
---
---
O
---
---
---
ML-80
I
I
---
O
---
---
---
I
O
---
---
---
---
HS20
I
I
---
O
---
---
---
I
O
---
---
---
---
H20
I
I
---
O
---
---
---
I
O
---
---
---
---
Spec. Veh.
I
I
---
O
---
---
---
I
O
---
---
---
—
P-82
B.3 - 6
DM-4, Section 3 - Loads and Load Factors
September 2007
Table D3.4.1.1P-1 - Load Factors and Live Load Vehicles for Steel Girders (Continued)
Notes:
1
Applicable when DL/LL ratio exceeds 7.0
A load factor of unity is applied to permanent loads for the fatigue and deflection limit state only when specified
3
DC load factor also used for barrier loads
4
DW load factor also used for utility loads
5
All loads applied to non-composite section for non-composite girders (Live loads are applied to the n section for steel)
6
Load combination for rating only
7
This row lists the typical design vehicle to be used for each load combination
8
The reduced load factor for LL with PL (see D3.4.1)
9
Design live load N/A for uncured slab check
2
Permanent Loads for Girder Programs
Load
Steel
Section Properties5
Steel
DC1
γGR
γSLAB
γSLAB
Girder
Slab
Haunch
Nc
nc
nc
DC2
γDC2
Barrier
3n
DW
γFWS
FWS
3n
B.3 - 7
September 2007
Table D3.4.1.1P-2 - Load Factors and Live Load Vehicles for Prestressed Concrete Girders
Load Combination
STR I
STR IP
8
6
STR IA
STR II
SERV I (P/S
compr. chk.)
SERV III (P/S
tension chk.
w/o
PL
with
PL
w/o
PL
with
PL
SERV IIIA
(Mr @ 0.9 fy
chk.)
SERV IIIB
(PennDOTcracking chk.)
FATIGUE1
DEFL1
γDC2
1.25,
0.90
1.25,
0.90
1.25,
0.90
1.25,
0.90
1.00
1.00
1.00
1.00
1.00
1.00
---
---
γDW3
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.00
1.00
1.00
1.00
1.00
1.00
---
---
γLL
1.75
1.35
1.10
1.35
1.00
0.80
0.804
0.659
1.00
1.00
0.75
1.00
γPL
—
1.75
---
---
---
1.00
---
1.00
---
---
---
---
γCR,SH
0.5
0.5
0.5
0.5
1.00
1.00
1.00
1.00
---
---
---
---
Design
LL
Veh.7
PHL-93
PHL-93
PHL-93
Permit
(P-82)
Controlling
PHL-93 or
P-82
Controlling
PHL-93 or P82
HS20-9.0
PennDOT
Defl. Trk.
PHL-93
Rating
Vehicle
PHL-93
PHL-93
I
I
O
---
I
I
O
---
---
---
---
---
---
O
---
---
O
---
---
---
ML-80
I
I
---
O
I
I
O
---
---
---
HS20
I
I
---
O
I
I
O
---
---
---
H20
I
I
---
O
I
I
O
---
---
---
Spec.
Veh.
I
I
---
O
I
I
O
---
---
—
P-82
B.3 - 8
September 2007
Table D3.4.1.1P-2 - Load Factors and Live Load Vehicles for Prestressed Concrete Girders (Continued)
Notes:
1
DC load factor also used for barrier loads
3
4
For rating vehicles, the live load for Service III is to be taken as 1.0 (γ = 0.80 for PHL-93 only)
5
All loads applied to non-composite section for non-composite girders
(Live loads are applied to the n section for P/S composite girders. For P/S, live load stresses can be based on transformed strands)
6
7
8
9
For rating vehicles (other than PHL-93), the live load factor for Service III is to be taken as 0.80 for the pedestrian load case
2
Permanent Loads for Girder Programs
Load
P/S
Section Properties5
P/S
DC1
γGIR
γSLAB
γSLAB
γID
γED
Girder
Slab
Haunch
Int. Dia.
Ext. Dia.
nc
nc
nc
nc
nc
DC2
γDC2
Barrier
n
DW
γFWS
FWS
n
B.3 - 9
September 2007
Table D3.4.1.1P-3 - Load Factors and Live Load Vehicles for Abutment/Retaining Walls
Load Combination
SERV I
STR I
STR IP
STR II
STR III
STR V
EXTREME I2
EXTREME II3
Min. γ for
Const. Case
(Strength)5
γ for
consolidation/second
ary settlement
γDC
1.00
1.25, 0.90
1.25, 0.90
1.25, 0.90
1.25, 0.90
1.25, 0.90
1.25, 0.90
1.25, 0.90
1.25, 0.90
1.00
γDW
1.00
1.50, 0.65
1.50, 0.65
1.50, 0.65
1.50, 0.65
1.50, 0.65
1.50, 0.65
0.00
---
1.00
γEV
1.00
γEV
γEV
γEV
γEV
γEV
γEV
γEV
γEV
1.00
γEH
1.00
γEH
γEH
γEH
γEH
γEH
0.00
γEH
γEH
1.00
γES4
1.00
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.00
γLS4
1.00
1.75
1.35
1.35
0.00
1.35
γEQ
0.50
1.50
0.00
1.00, 0
1.75, 0
1.35, 0
1.35, 0
0.00
1.35, 0
γEQ, 0
0.00
---
0.00
γPL
0.00
0.00
1.75, 0.00
0.00
0.00
0.00
0.00
0.00
---
0.00
γWS
0.3
0.00
0.00
0.00
1.40
0.40
0.00
0.00
---
0.00
γWL
1.0
0.00
0.00
0.00
0.00
1.0
0.00
0.00
---
0.00
γWA
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
γBR
1.0
1.75
1.35
1.35
0.00
1.35
γEQ
0.00
---
0.00
γCE
1.0
1.75
1.35
1.35
0.00
1.35
γEQ
0.00
---
0.00
γFR
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
---
0.00
γTU
1.0
0.5
0.5
0.5
0.5
0.5
0.00
0.00
---
0.00
γEQ
0.00
0.00
0.00
0.00
0.00
0.00
1.0
0.00
---
0.00
γCT
0.00
0.00
0.00
0.00
0.00
0.00
0.00
1.0
---
0.00
PHL-93
PHL-93
PHL-93
P-82
---
---
PHL-93
---
---
—
γLLIM1
Design LL
Vehicle
B.3 - 10
B.3 - 11
September 2007
September 2007
Table D3.4.1.1P-3 - Load Factors and Live Load Vehicles for Abutment/Retaining Walls (Continued)
Notes:
1
For a negative reaction on an abutment (uplift), use the maximum load factor
For the seismic load case, EH loads (normal lateral earth pressure) replaced by EQ soil loads. γ EQ for live loads = 0.0.
3
Parapet collision force, CT.
4
All lateral loads and their vertical components are maximized.
5
For evaluation of the temporary construction stages using the Strength Limit states (see D11.6.1.2), use the greater of the γ noted under Construction Case column or under the
given Strength Limit State column.
2
Abutment/Retaining Wall
Earth Load Factors
Maximum
Minimum
γEV
1.35
1.00
γEH
1.50
(4)
B.3 - 12
September 2007
Table D3.4.1.1P-4 - Load Factors and Live Load Vehicles for Box Culverts
Load Combination
SERV I
STR I
STR IA
STR II
FATIGUE1
Min. γ for
Const. Case
(Strength)4
γDC
1.00
1.25, 0.90
1.25, 0.90
1.25, 0.90
---
1.25, 0.90
γDW
1.00
1.50, 0.65
1.50, 0.65
1.50, 0.65
---
---
γEV
1.00
γEV
γEV
γEV
—
γEV
γEH
1.00
γEH
γEH
γEH
—
γEH
γES3
1.00
1.50, 0.75
1.50, 0.75
1.50, 0.75
---
1.50,0.75
γLS
1.00, 0
1.75, 0
1.10, 0
1.35, 0
---
1.50, 0
γLLIM
1.00, 0
1.75, 0
1.10, 0
1.35, 0
0.75
---
PHL-93
PHL-93
PHL-93
P-82
HS20-9.0
---
Design LL Vehicle
Rating Vehicle
2
PHL-93
---
I
O
---
---
---
P-82
---
---
---
O
---
---
ML-80
---
I
---
O
---
---
HS20
---
I
---
O
---
---
H20
---
I
---
O
---
---
Spec. Veh.
—
I
---
O
---
---
Notes:
1
Fatigue load factor should be factored by PTF. A load factor of unity is applied to permanent loads for the fatigue limit state only when specified.
Rating applicable for box culverts only
3
Minimum ES of 0.50 applies for top slabs of box culverts
4
See A3.4.2, Load Factors for Construction Loads
2
B.3 - 13
Box Culvert
Earth Load Factors
Maximum
Minimum
γEV
1.30
0.90
γEH
1.35
0.90*
*
Use 0.50 minimum for culvert top slab
B.3 - 14
September 2007
September 2007
Table D3.4.1.1P-5 - Load Factors and Live Load Vehicles for Steel Floorbeams
Load Combination
STR IA
STR II
STR III
STR V
SERV II
SERV
IIA5
SERV
IIB
FATIGUE1
DEFL1
CONST/
UNCURED
SLAB8
1.25,
0.90
1.25,
0.90
1.25,
0.90
1.25,
0.90
1.25,
0.90
1.00
1.00
1.00
---
---
1.25
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.00
1.00
1.00
---
---
1.50, 0.65
γLL
1.75
1.35
1.10
1.35
---
1.35
1.30
1.00
1.00
0.75
1.00
1.50
γPL
—
1.75
---
---
---
---
---
---
---
---
---
---
γWS
—
---
—
—
1.40
0.40
---
---
---
---
---
1.25
PHL-93
PHL-93
PHL-93
Permit
(P-82)
---
PHL-93
PHL-93
PHL-93
Permit
(P-82)
HS20-9.0
PennDOT
Defl. Trk.
User Def.
STR I
STR IP
γDC2
1.25,
0.90
γDW3
Design LL
Veh.6
7
5
Rating Veh.
PHL-93
I
I
O
---
---
---
I
O
---
---
---
---
---
---
---
O
---
---
---
---
O
---
---
---
ML-80
I
I
---
O
---
---
I
O
---
---
---
---
HS20
I
I
---
O
---
---
I
O
---
---
---
---
H20
I
I
---
O
---
---
I
O
---
---
---
---
Spec. Veh.
I
I
---
O
---
---
I
O
---
---
---
—
P-82
B.3 - 15
September 2007
Table D3.4.1.1P-5 - Load Factors and Live Load Vehicles for Steel Floorbeams (Continued)
Notes:
1
DC load factor also used for barrier loads, sidewalk, median barrier, railings, etc.
3
4
All loads applied to non-composite section for non-composite floorbeams (Live loads are applied to the n section for steel)
5
6
7
8
Live load N/A for uncured slab check
2
Permanent Loads for Floorbeam Programs
Load
Section Properties4
Steel
Steel
DC1
γFLBM
γSLAB
γSLAB
Floorbeam
Slab
Haunch
Nc
nc
nc
DC2
γDC2
Barrier, Sidewalk,
Median Barrier,
Railings, etc.
3n
DW
γFWS
FWS
3n
B.3 - 16
September 2007
Table D3.4.1.1P-6 - Load Factors and Live Load Vehicles for Steel Trusses
Load Combination
STR I
STR
IP8
STR
IA5
STR
II
STR
III
STR
IV1
STR V
EXT.
EVEN
T
III
EXT.
EVEN
T
IV
SERV
II
SERV
IIA5
SER
V IIB
FATIGUE2
DEFL2
CONST
γDC3
1.25,
0.90
1.25,
0.90
1.25,
0.90
1.25,
0.90
1.25,
0.90
1.5
1.25,
0.90
1.25,
0.90
1.05,
0.95
1.00
1.00
1.00
---
---
1.25
γDW4
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.50,
0.65
1.05,
0.90
1.00
1.00
1.00
---
---
1.50,
0.65
γLL
1.75
1.35
1.10
1.35
---
---
1.35
1.30
1.15
1.30
1.00
1.00
0.75
1.00
1.50
γPL
---
1.75
---
---
---
---
---
1.10
1.05
---
---
---
---
---
---
γWS
---
—
---
—
1.40
---
0.40
---
---
---
---
---
---
---
1.25
PHL93
PHL93
PHL93
Permi
t
(P-82)
---
---
PHL93
PHL93
P-82
PHL93,
P-82
PHL93
PHL93
Permi
t
(P82)
HS20-9.0
PennDO
T
Defl.
Trk.
User
Def.
Design
LL
Veh.6
Rating
Veh.
PHL-93
I
I
O
---
---
---
---
---
---
I
O
---
---
---
---
---
---
---
O
---
---
---
---
---
---
---
O
---
---
---
ML-80
I
I
---
O
---
---
---
---
---
I
O
---
---
---
---
HS20
I
I
---
O
---
---
---
---
---
I
O
---
---
---
---
H20
I
I
---
O
---
---
---
---
---
I
O
---
---
---
---
Spec.
Veh.
I
I
---
O
---
---
---
---
---
I
O
---
---
---
—
P-82
B.3 - 17
Table D3.4.1.1P-6 - Load Factors and Live Load Vehicles for Steel Trusses (Continued)
Notes:
1
Applicable when DL/LL ratio exceeds 7.0
3
DC load factor also used for barrier loads, sidewalk, median barrier, railings, deck, stringers, truss floorbeams, wind and lateral bracing, etc.
4
5
6
7
2
B.3 - 18
September 2007
September 2007
SPECIFICATION
COMMENTARY
3.5 PERMANENT LOADS
C3.5.1
3.5.1 Dead Loads: DC, DW and EV
The normal density concrete and low density concrete
with densities of 2400 kg/m3 {0.150 kcf} and 1840 kg/m3
{0.115 kcf} respectively include an allowance for
reinforcement in the calculation of the density.
For concrete deck slabs, provisions must be made in the
design for the addition of a bituminous wearing surface at
some future time. Even in cases where the initial design
includes a bituminous surface, provision must be made for
an additional future wearing surface since the original
bituminous material is not always stripped off before the
new surface is added.
For structures under fill, the additional dead load
associated with a future wearing surface is insignificant
when compared with other contributions to the dead load.
Therefore, in this case, no allowance for future wearing
surface is necessary.
It is recognized that permanent metal deck forms are
available for which the surface density is less than 75 kg/m2
{0.015 ksf}; however, the minimum design load should be
retained at this level.
Lightweight forms may be
advantageous in certain situations, such as rehabilitation,
and should be evaluated on a case-by-case basis.
The acceleration due to gravity (g) shall be taken as
9.81 m/s2 for use in determining force effects from mass.
In addition to the weight of the deck slab, the design
dead load shall include provisions for a future wearing
surface with a surface area density of 150 kg/m2 {0.030 ksf}
on the deck slab between the curbs. This load shall be
considered for all deck slabs, including decks with a
bituminous wearing surface, but not for structures under fill.
For decks formed using permanent metal deck forms, an
additional dead load shall be included based on a surface
density of 75 kg/m2 {0.015 ksf} which takes into account
the weight of the form, plus the weight of the concrete in the
valleys of the forms.
In Table A3.5.1-1, replace the low density concrete
value of 1775 kg/m3 {0.110 kcf} with 1840 kg/m 3 {0.115
kcf}. Also in Table A3.5.1-1, delete the "sand-low-density"
concrete value. For use of low density concrete with
densities different than 1840 kg/m3 {0.115 kcf}, see
D5.4.2.1 and DC5.4.2.1.
3.5.1.1P APPLICATION OF DEAD LOAD ON GIRDER
AND BOX BEAM STRUCTURES
The provisions in this article apply to superstructure
types, a, b, c, f, g, h, k and l given in Table A4.6.2.2.1-1.
For composite adjacent and spread beams, the barrier
(single barrier) load shall be equally distributed to the
nearest three and two beams, respectively, when the barriers
are placed after slab has hardened.
The dead load of items, such as fencing and sound
barriers, if placed after the slab has hardened, shall be
distributed to girder as described above.
Sidewalk dead load shall be distributed to a girder using
the lever rule.
For noncomposite girders, the barrier load shall be
distributed solely to the fascia girder.
The future wearing surface shall be distributed equally
among all girders.
3.6 LIVE LOADS
3.6.1 Gravity Loads: LL, PL
3.6.1.1 VEHICULAR LIVE LOAD
3.6.1.1.2 Multiple Presence of Live Load
C3.6.1.1.2
Delete the last sentence of the second paragraph of
A3.6.1.1.2.
Delete the third through seventh paragraphs of
AC3.6.1.1.2.
B.3 - 19
SPECIFICATION
September 2007
COMMENTARY
3.6.1.2 DESIGN VEHICULAR LIVE LOAD
3.6.1.2.1 General
C3.6.1.2.1
A3.6.1.2.1.
The vehicular live loading on the roadways of bridges
or incidental structures, designated PHL-93, shall consist of
a combination of the:
The following shall supplement AC3.6.1.2.1.
At this time, the Department makes no exceptions to
the requirements for application of PHL-93 vehicular live
load for bridges on low volume roads.
design truck or design tandem, and
design lane load,
as given in A3.6.1.2 and D3.6.1.2.
3.6.1.2.3 Design Tandem
Modify A3.6.1.2.3 so that weight of each axle is
increased from 110 kN to 140 kN {25 kips to 31.25 kips}.
C3.6.1.2.5
The area load applies only to the design truck and tandem.
For other design vehicles, the tire contact area should be
determined by the engineer.
As a guideline for other truck loads, the tire area in mm2
{in2} may be calculated from the following dimensions:
Metric Units:
Tire width = P/0.142
Tire length = 165γ(1+IM/100)
U.S. Customary Units:
Tire width = P/0.8
Tire length = 6.4γ(1+IM/100)
where:
γ
B.3 - 20
=
load factor, as given in A3.4.1 and D3.4.1, except
for buried structures where the load factor shall be
1.35
IM =
dynamic load allowance percent
P
wheel load
72.5 kN {16 kips} for the design truck, 70 kN
{15.625 kips} for the design tandem and 60 kN
{13.5 kips} for the P-82
=
=
SPECIFICATION
September 2007
COMMENTARY
A constant value of γ was chosen for buried structures
to simplify the analysis at strength and service limit states.
PennDOT has conducted a study to ensure that the use of a
constant load factor has a negligible effect.
3.6.1.2.6 Distribution of Wheel Loads through Earth Fills
C3.6.1.2.6
A3.6.1.2.6.
Where the depth of fill is less than 600 mm {2 ft.}, live
loads shall be distributed to the top slabs of culverts as
specified in D4.6.2.12P.
The following shall replace the second paragraph of
A3.6.1.2.6.
In lieu of a more precise analysis, or the use of other
acceptable approximate methods of load distribution
permitted in Section 12, where the depth of fill is 600 mm
{2 ft.} or greater, wheel loads may be considered to be
uniformly distributed over a rectangular area with sides
equal to the dimension of the tire contact area, as specified
in A3.6.1.2.5, and increased by either 1.15 times the depth
of the fill in select granular backfill, or the depth of the fill
in all other cases. The provisions of A3.6.1.1.2 and
A3.6.1.3 shall apply.
A3.6.1.2.6.
Where live load and impact moment in concrete slabs,
based on the distribution of wheel load through earth fills,
exceeds the live load and impact moment calculated
according to A4.6.2.1 and D4.6.3.2, the latter moment shall
be used.
Traditionally, the effect of fills less than 600 mm {2 ft.}
deep on live load has been ignored. Research (McGrath, et
al. 2004) has shown that in design of box sections allowing
distribution of live load through fill in the direction parallel
to the span provides a more accurate design model to predict
moment, thrust, and shear forces. Provisions in D4.6.2.12P
provide a means to address the effect of shallow fills.
3.6.1.2.7P Design Permit Load (P-82)
The weights and spacings of axles and wheels for the
Permit Load (P-82) shall be as specified in Figure 1. A
dynamic load allowance shall be considered as specified in
D3.6.2 and A3.6.2.
NOTE:
P-82 width is the same as the Design Truck.
Transverse wheel location is the same as Design Truck.
B.3 - 21
SPECIFICATION
September 2007
COMMENTARY
Figure 3.6.1.2.7P-1 - Pennsylvania Permit Load (P-82) 910 kN {102 tons}, 8 Axle
Axles which do not contribute to the extreme force
effect under consideration shall be neglected.
For multi-girder superstructures design, the permit load
shall be in one lane or in multiple lanes whichever is the
controlling case.
For superstructure with girder-floorbeam-stringer
systems and substructure components designs, the permit
load shall be in one lane or in one lane with PHL-93 loading
in adjacent lanes, whichever is the controlling case.
3.6.1.2.8P Maximum Legal Load (ML-80)
C3.6.1.2.8P
The weights and spacings of axles and wheels for the
Maximum Legal Load (ML-80) shall be as specified in
Figure 1. The ML-80 truck is used for rating.
The ML-80 is not considered a notional load.
Therefore, all of the axles shall be considered when
determining force effects.
NOTE: ML-80 width is the same as the design truck.
Transverse wheel location is the same as design
truck.
Figure 3.6.1.2.8P-1 - Pennsylvania Maximum Legal Load
(ML-80) 335.7 kN {37.74 tons}, 4 Axle
3.6.1.3 APPLICATION OF DESIGN VEHICULAR LIVE
LOADS
3.6.1.3.1 General
C3.6.1.3.1
Unless otherwise specified, the extreme force effect
shall be taken as the larger of the following:
Delete the second and third sentences of the third
paragraph of AC3.6.1.3.1.
The BXLRFD program does not consider the effect of
two design trucks, since the minimum distance between the
two design trucks is 15 000 mm {50 ft.} which is at the
upper limit of a twin cell culvert. The effects of two
tandems are considered for a twin cell box culvert in the
BXLRFD program.
the effect of the design tandem combined with the
effect of the design lane load, or
the effect of one design truck with the variable axle
spacing specified in A3.6.1.2.2, combined with the
effect of the design lane load, and
for the negative moment between points of dead load
contraflexure, the effect of two design trucks spaced a
minimum of 15 000 mm {50 ft.} between the lead axle
of one truck and the rear axle of the other truck,
B.3 - 22
SPECIFICATION
September 2007
COMMENTARY
combined with the effect of the design lane load; the
distance between the 145 kN {32 kips} axles of each
truck shall be taken as 4300 mm {14 ft.}.
For the reaction at interior piers only, 90% of the effect
of two design trucks spaced a minimum of 15 000 mm
{50 ft.} between the lead axle of one truck and the rear
axle of the other truck, combined with 90% of the effect
of the design lane load. The distance between the 145
kN {32 kips} axles of each truck shall be taken as 4300
mm {14 ft.}.
For the negative moment between points of dead load
contraflexure, the effect of two tandems with axle
weights of 110 kN {25 kips} spaced from 8000 mm to
12 000 mm {26 ft. to 40 ft.} apart, combined with the
effect of the design lane load.
For the reaction at interior piers only, 100% of the
effect of two tandems with axle weights of 110 kN {25
kips} spaced from 8000 mm to 12 000 mm {26 ft. to 40
ft.} apart combined with the effect of the design lane
load.
Axles which do not contribute to the extreme force
effect under consideration shall be neglected.
Both the design lanes and the position of the 3000 mm
{10 ft.} loaded width in each lane shall be positioned to
produce extreme force effects. The design truck or tandem
shall be positioned transversely such that the center of any
wheel load is not closer than:
for the design of the deck overhang - 300 mm {1 ft.}
from the face of the curb or railing, and
for the design of all other components - 600 mm {2 ft.}
from the edge of the design lane.
Unless otherwise specified, the lengths of design lanes, or
parts thereof, which contribute to the extreme force effect
under consideration, shall be loaded with the design lane
load.
3.6.1.3.2 Loading for Live Load Deflection Evaluation
The deflection should be taken as 125% of the larger of:
that resulting from the effect of one design truck with
the variable axle spacing specified in A3.6.1.2.2,
that resulting from the effect of 25% of one design
truck with the variable axle spacing specified in
A3.6.1.2.2, combined with the effect of the design lane.
B.3 - 23
C3.6.1.3.2
The LRFD live load deflection criteria was developed
such that deflections would be roughly equivalent to those
produced by a HS20 vehicle. A 25% increase is specified to
be consistent with the Department's past use of the HS25
vehicle for computing deflections.
September 2007
SPECIFICATION
COMMENTARY
3.6.1.3.3 Design Loads for Decks, Deck Systems, and the
Top Slab of Box Culverts
C3.6.1.3.3
Replace the three bullets of the second paragraph of
A3.6.1.3.3 with the following.
Where the slab spans primarily in the transverse
direction, only the axles of the design truck of
A3.6.1.2.2 or design tandem of D3.6.1.2.3 shall be
applied to the deck slab or top slab of box culverts.
Add the following after the second paragraph of
AC3.6.1.3.3.
The design truck and tandem without the lane load and
with a multiple presence factor of 1.2 results in factored
force effects that are similar to the factored force effects
using the Standard Specification for typical span ranges of
box culverts.
Where the slab spans primarily in the longitudinal
direction:
o
o
For top slabs of box culverts of all spans
and for all other cases, including slab-type
bridges where the span does not exceed
4600 mm {15.0 ft.}, only the axle loads of
the design truck or design tandem of
A3.6.1.2.2 and D3.6.1.2.3, respectively,
shall be applied.
For all other cases, including slab-type
bridges (excluding top slabs of box
culverts) where the span exceeds 4600
mm {15.0 ft.}, all of the load shall be
applied.
Replace the third paragraph of A3.6.1.3.3 with the
following:
Where the refined methods are used to analyze decks,
force effects shall be determined on the following basis:
Where the slab spans primarily in the transverse
direction, only the axles of the design truck of
A3.6.1.2.2 or design tandem of D3.6.1.2.3 shall be
applied to the deck slab.
Where the slab spans primarily in the longitudinal
direction (including slab-type bridges), all of the
loads specified in D3.6.1.2 shall be applied.
3.6.1.3.4 Deck Overhang Load
The deck overhang load shall be as given in D3.6.1.3.1.
Also, the ultimate strength of the deck section shall be
greater than the ultimate strength of the barrier, see Section
A13 and its Appendix. Horizontal loads on the overhang
resulting from vehicle collision with barriers shall be in
accordance with the provisions of Section A13 and its
Appendix.
B.3 - 24
C3.6.1.3.4
The deck overhang slab provided in BD-601M has been
designed for the vertical design loads (D3.6.1.3.1) or a
strength greater than the applied forces transmitted to the
overhang when the barrier is subjected to the maximum
collision force it can resist (Section A13) whichever is
greater. The ultimate strength of the barrier used in the
design of the overhang was based on the Department’s
Typical Barrier (see Standard Drawing BD-601M) which
placed greater demand on the deck overhang than the other
Department barriers.
SPECIFICATION
September 2007
COMMENTARY
3.6.1.4 FATIGUE LOAD
3.6.1.4.2 Frequency
The following shall replace Table A3.6.1.4.2-1.
Table 3.6.1.4.2-1 – Fraction of Truck Traffic in a Single
Lane, p
Number of Lanes
Available to Trucks
p
1
1.00
2 or more
0.85
3.6.1.5 RAIL TRANSIT LOAD
3.6.1.5.1P General
Live loads for rail traffic shall use a combination of
axle loads and axle spacings represented by the Cooper E80
loading, as shown in Figure 1.
Figure 3.6.1.5.1P-1 - Wheel Spacing for Cooper E80 Design Loading (Load/Axle)
3.6.1.5.2P Distribution of Rail Transit Loads Through Earth
Fill
The load intensity, Wl, on a buried structure due to rail
transit loading shall be determined using the following
relationship:
B.3 - 25
SPECIFICATION
Wl = C Po Bc (1 + If)
September 2007
COMMENTARY
(3.6.1.5.2P-1)
Refer to Table 1 for values of C. The series of axle
loads and spacing shall be converted into a uniform load at
the bottom of the railroad ties. The loading, P o, at the base
of the ties shall be represented by a ground pressure of
97 MPa {2025 ksf}, which represents the locomotive drivewheel (four at 360 kN {80 kips}) loading distributed over an
area 2400 mm by 6100 mm {8 ft. by 20 ft.} and a track
structure loading of 3 kN/m {0.2 kip/ft}. The impact factor,
If, shall range from 40% at zero cover to 0% at 3000 mm
{10 ft.} of cover.
The live load and the dead load, including the impact
factor, for a Cooper E80 loading can be determined from
Figure 1. To obtain the live load per linear meter, multiply
the unit load from Figure 1 by the outside horizontal span of
the pipe, Bc.
B.3 - 26
Figure 3.6.1.5.2P-1 - Live and Dead Loads on Pipe Installed Under Railroads (ACPA, 1981)
B.3 - 27
September 2007
Table 3.6.1.5.2P-1 - Values of Load Coefficient (C) for Concentrated and Distributed Superimposed Loads Vertically Centered Over Culvert (ASCE, 1969)
D
2H
or
M
L
or
2H
2H
Bc
2H
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.2
1.5
2.0
5.0
0.1
0.019
0.037
0.053
0.067
0.079
0.089
0.097
0.103
0.108
0.112
0.117
0.121
0.121
0.128
0.2
0.037
0.072
0.103
0.131
0.155
0.174
0.189
0.202
0.211
0.219
0.229
0.238
0.211
0.218
0.3
0.053
0.103
0.149
0.190
0.221
0.252
0.274
0.292
0.306
0.318
0.333
0.345
0.355
0.360
0.4
0.067
0.131
0.190
0.241
0.281
0.320
0.349
0.373
0.391
0.405
0.425
0.440
0.454
0.460
0.5
0.079
0.155
0.224
0.284
0.336
0.379
0.414
0.441
0.463
0.484
0.505
0.525
0.540
0.548
0.6
0.089
0.171
0.252
0.320
0.379
0.428
0.467
0.499
0.524
0.544
0.572
0.596
0.613
0.624
0.7
0.097
0.189
0.274
0.349
0.414
0.467
0.511
0.516
0.584
0.597
0.628
0.650
0.674
0.688
0.8
0.103
0.202
0.292
0.373
0.441
0.499
0.546
0.581
0.615
0.639
0.674
0.703
0.725
0.740
0.9
0.108
0.211
0.306
0.391
0.463
0.524
0.574
0.615
0.647
0.673
0.711
0.742
0.766
0.784
1.0
0.112
0.219
0.318
0.405
0.481
0.544
0.597
0.639
0.673
0.701
0.740
0.774
0.800
0.816
1.2
0.117
0.229
0.333
0.425
0.505
0.572
0.628
0.674
0.711
0.740
0.783
0.820
0.819
0.868
1.5
0.121
0.238
0.345
0.440
0.525
0.596
0.650
0.703
0.742
0.774
0.820
0.861
0.891
0.916
2.0
0.121
0.211
0.355
0.454
0.540
0.613
0.674
0.725
0.766
0.800
0.819
0.894
0.930
0.956
*Influence coefficients for solution of Holl's and Newmark's integration of the Boussinesq equation for vertical stress
B.3 - 28
SPECIFICATIONS
September 2007
COMMENTARY
3.6.1.6 PEDESTRIAN LOADS
The pedestrian load is distributed using the lever rule.
When pedestrian loads are to be considered, two
loading conditions shall be considered. The first loading
condition assumes that the sidewalk is not present (i.e., an
extended roadway surface and barrier would replace the
sidewalk area) and the bridge is used for vehicular live load
only. Under the second loading condition, the pedestrian
load is present and the vehicular live load is factored at a
reduced level. The Strength IP load combination was
developed for the second loading condition.
3.6.2 Dynamic Load Allowance: IM
3.6.2.1 GENERAL
C3.6.2.1
For permit loads, the static effect of the P-82 shall be
increased by a percentage not to exceed IM = 20%.
IM for deck design = 50%
The second to last paragraph in A3.6.2.1 which begins
"Dynamic load allowance need not..." shall be deleted.
Irregularities in decks such as potholes can result in
large localized impact effects. As a result, PennDOT
requires that the impact for decks be increased from 33% to
50% for decks. Other elements of the bridge structure
should not be greatly affected by high localized impact due
to dampening. The combination of 50% impact, the design
truck (former HS20 truck) and LRFD deck design criteria
will produce deck designs comparable to 30% impact, HS25
and past AASHTO deck design criteria.
3.6.2.1.1P Components for which IM is Applicable
The following components shall have the IM factor
included in the design:
all superstructure components including deck and deck
joints
pier caps and shafts
backwalls and pedestals of abutments
bearings, except for plain and reinforced elastomeric
bearings
For buried components covered in D12 and A12, see
D3.6.2.2. For wood component, see D3.6.2.3.
3.6.2.1.2P Components for which IM is Not Applicable
C3.6.2.1.2P
The following components shall not have the IM factor
included in the design:
The PAPIER program carries the live loads from the
pier cap through to the footing without the removal of the
effect of the dynamic load allowance (IM) input by the user.
This provides a consistent mathematical model throughout
the structure, where the moments, shears, and axial forces at
the bottom of the column are equal to those at the top of the
footing of the pier.
retaining walls not subject to vertical reactions from the
superstructure, including MSE walls
foundation components which are entirely below
B.3 - 29
SPECIFICATIONS
September 2007
COMMENTARY
ground level, including footings (except for frame and
box culverts where IM is applicable as per A3.6.2.2),
piles, caissons and pedestals
abutment stems
plain and reinforced elastomeric bearings
buried components with 2400 mm {8 ft.} or greater fill
above them (see A3.6.2.2)
The pedestrian load shall not have the IM factor
applied.
3.6.2.2 BURIED COMPONENTS
The following shall replace A3.6.2.2.
The dynamic load allowance for culverts and other
buried structures covered by Section 12, in percent, shall be
taken as:
Metric Units:
IM = 40 (1.0 - 4.1 x 10-4 DE) ≥ 0%
(3.6.2.2-1)
IM = 40 (1.0 - 0.125 DE) ≥ 0%
where:
DE =
the minimum depth of earth cover above the
structure (mm) {ft.}
Dynamic load allowance shall not be applied to foundation
pressures.
3.6.2.3 WOOD COMPONENTS
C3.6.2.3
Delete the second sentence of CA3.6.2.3.
For wood bridges and wood components of bridges, the
dynamic load allowance specified in A3.6.2.1 may be
reduced to 50 percent of the values specified for IM in Table
A3.6.2.1-1.
3.6.4 Braking Force: BR
C3.6.4
Dynamic load allowance is not applied to the braking
force.
B.3 - 30
The following shall supplement CA3.6.4.
LRFD analysis of the capacity of existing substructure
units on shorter span bridges may become problematic. Use
of the original design braking force, requiring approval of
the Chief Bridge Engineer, may be warranted for analysis of
these older structures.
SPECIFICATIONS
September 2007
COMMENTARY
Braking Force Factor Tables previously included in
DM-4 have been deleted since AASHTO 3rd edition added
lane loading.
The AASHTO 2nd edition used only axle loads, thus
Braking Force Factor was developed.
3.6.5 Vehicular Collision Force CT
3.6.5.3 Vehicular Collision with barriers
C3.6.5.3
For transverse vehicular collision loading transferred to
the substructure for u-wings and retaining walls, use a load
of 45 kN {10 kip} acting over 1.5 m {5 ft} length applied at
a distance equal to the height of the concrete barrier above
the top of the wall.
The following shall supplement AC3.6.5.3.
The transverse vehicular collision loading of 45 kN {10
kip} acting over 1.5 m {5 ft} may be distributed down to the
footing at a 1:1 slope. Adjacent to open joints, this load
may only be distributed in one direction which will usually
be the controlling condition. Distributing the load in one
direction is conservative for footing designs, since the
footings are continuous at open joints.
3.8 WIND LOAD: WL AND WS
3.8.1 Horizontal Wind Pressure
3.8.1.2 WIND PRESSURE ON STRUCTURES: WS
3.8.1.2.1 General
C3.8.1.2.1
B.3 - 31
SPECIFICATIONS
September 2007
COMMENTARY
AC3.8.1.2.1.
If approved by the Chief Bridge Engineer, wind tunnel
tests may be used to provide more precise estimates of wind
pressures. Such testing should be considered where wind is
a major design load.
C3.8.1.2.2
The columns referenced in A3.8.1.2.2 are columns in
arch bridges not columns in substructure units.
3.8.3 Aeroelastic Instability
3.8.3.4 WIND TUNNEL TESTS
If approved by the Chief Bridge Engineer,
representative wind tunnel tests may be used to satisfy the
requirements of A3.8.3.2 and A3.8.3.3.
3.9 ICE LOADS: IC
3.9.1 General
C3.9.1
The forces due to ice shall be applied at the average
elevation of the highest expected water elevation and the
normal water elevation.
The PAPIER program uses a default ice thickness of
150 mm {6 in.} and a default ice crushing strength of
2.75 MPa {58 ksf}.
3.9.5 Vertical Forces due to Ice Adhesion
C3.9.5
The following shall replace A3.9.5.
The vertical force on a bridge pier due to rapid water
level fluctuation shall be taken as:
Metric Units:
for a circular pier, in N:
0.3 t 2 + 0.0169 R t 1.25
for an oblong pier, in N:
0.3 t 2 + 0.0169 R t 1.25 + 2.3*10 -3 L t 1.25
for a circular pier, in kips:
6.27 t 2 + 1.48 R t 1.25
for an oblong pier, in kips
6.27 t 2 + 1.48 R t 1.25 + 0.2 L t 1.25
where:
B.3 - 32
Delete AC3.9.5.
SPECIFICATIONS
September 2007
COMMENTARY
t
=
ice thickness (mm) {ft.}
R
=
radius of circular pier (mm) {ft.} or approximated
end of oblong pier
L
=
perimeter of pier excluding half circles at ends of
oblong pier (mm) {ft.}
3.10 EARTHQUAKE EFFECTS: EQ
C3.10.1 General
Minimize bridge skew as much as and whenever
possible. It is well known that skewed structures perform
poorly in seismic events when compared to the performance
of normal or non-skewed structures.
These specifications present seismic design and
construction requirements applicable to the majority of
highway bridges to be constructed in the United States.
Bridges not covered by these provisions probably constitute
5 to 15 percent of the total number of bridges designed.
The Project Engineering Panel (PEP) of the Applied
Technology Council (ATC) has decided that special seismic
design provisions are not required for buried structures. It
was recognized by the PEP, however, that this decision may
need to be reconsidered as more research data on the seismic
performance of this type of structure become available.
These specifications specify minimum requirements.
More sophisticated design or analysis techniques may be
utilized if deemed appropriate by the Design Engineer.
For bridge types not covered by these specifications, the
following factors should be considered.
(a) The recommended elastic design force levels of the
specifications should be applicable because force
levels are largely independent of the type of bridge
structure. It should be noted that the elastic design
force levels of the specifications are part of a
design philosophy described in Chapter 1 of
FHWA Research Report FHWA-IP-87-6. The
appropriateness of both the design force levels and
the design philosophy must be assessed before they
are used for bridges that are not covered by these
specifications.
(b) A multi-mode dynamic analysis
considered, as specified in A4.7.4.3.3
should
be
(c) Design displacements are as important as design
forces; when possible, the design methodology
should consider displacements arising from the
effects discussed in DC4.7.4.4.
B.3 - 33
SPECIFICATIONS
September 2007
COMMENTARY
(d) If a design methodology similar to that used in
these specifications is
deemed desirable, the
design requirements contained herein should be
used to ensure compliance with the design
philosophy.
Use caution when referencing the flowcharts contained
in the Appendix to A3. They reference AASHTO LRFD
Specifications which may be modified by the Design
Manual, Part 4.
3.10.2 Acceleration Coefficient
The acceleration coefficient, A, to be used in the
application of these provisions shall be determined from the
County designations given in Figure 1.
B.3 - 34
SPECIFICATIONS
September 2007
COMMENTARY
Figure 3.10.2-1 - Acceleration Coefficients for Pennsylvania Counties
3.10.7 Response Modification Factors
3.10.7.1 GENERAL
A3.10.7.1.
Do not apply the response modification factors to single
span structures or to structures in Seismic Zone 1.
The use of R-factors requires the reinforcing details
meet the requirements of this document and the LRFD
Specifications for consideration of structural ductility. All
R-factors from Table A3.10.7.1-1 must be reduced to 1.0 if
the details do not meet the specification requirements,
unless tests have been done to indicate otherwise.
3.10.9 Calculation of Design Forces
3.10.9.2 SEISMIC ZONE 1
C3.10.9.2
Prior to the redesign phase of the PEP project, the PEP
thought that the design of connections for wind forces
would be satisfactory for anticipated seismic forces for
bridges in Seismic Zone 1. However, when the magnitude
of the wind and seismic forces were compared for six
bridges, it was found in almost all cases that, for an
B.3 - 35
SPECIFICATIONS
September 2007
COMMENTARY
acceleration coefficient of 0.10, seismic forces were greater
than wind forces. In some cases the difference was
significant. Hence it was deemed necessary to include the
requirement of this section for the design of the connections.
The requirement is simple and somewhat conservative,
especially for more flexible bridges, since the forces are
based on the maximum elastic response coefficient. If the
design forces are difficult to accommodate, it is
recommended that Seismic Zone 2 analysis and design
procedures be used.
3.10.9.3 SEISMIC ZONE 2
C3.10.9.3
The seismic design forces specified for bridges in
Seismic Zone 2 are intended to be relatively simple, but
consistent with the overall design concepts and
methodology. Inherent in any simplification of a design
procedure, however, is a degree of conservatism; for
Seismic Zone 2 this occurs in the determination of the
design forces for the foundations and connections to
columns. If these forces appear to be excessive, more
refined methods may be used. An acceptable approach
would be to use the methods suggested for Seismic Zones 3
and 4 in the LRFD Specifications. For such a refinement,
the foundations and connections to columns are designed for
the maximum forces that a column can transmit to these
components. In some cases, these may be considerably less
than the design forces specified in this section.
This section specifies the design forces for the
structural components of the bridge. In the first step, the
elastic forces of Load Cases 1 and 2 of A3.10.8 are divided
by the appropriate R-factors of A3.10.7. These forces are
combined with those from other loads in load combination,
Extreme Event I. Each component shall be designed to
resist two seismic load combinations, one including Load
Case 1 and the other including Load Case 2. Each load case
incorporates different proportions of bi-directional seismic
loading. This may be important for some components (e.g.,
bi-axial design of columns) and unimportant for others. In
the design loads for each component, the sign of the seismic
forces and moments obtained from A3.10.8 can be taken as
either positive or negative. The sign of the seismic force or
moment that gives the maximum magnitude for the design
force (either positive or negative) shall be used.
This section also specifies the design forces for
foundations which include the footings, pile caps, and piles.
The design forces are essentially twice the seismic design
forces of the columns. This will generally be conservative
and was adopted to simplify the design procedure for
bridges in Seismic Zone 2. However, if seismic forces do
not govern the design of columns and piers, there is a
possibility that during an earthquake the foundations will be
subjected to forces larger than the design forces. This will
occur if the columns remain elastic throughout the duration
B.3 - 36
SPECIFICATIONS
September 2007
COMMENTARY
of the seismic ground motion. Thus, for essential bridges in
Seismic Zone 2, consideration should be given to the use of
more refined methods in the design of the foundation. An
acceptable approach would be to use the methods suggested
for Seismic Zones 3 and 4 in the LRFD Specifications. It
should be noted that ultimate soil and pile strengths are to be
used with the specified foundation seismic design forces.
3.10.9.5 LONGITUDINAL RESTRAINERS
Restrainers may only be used with the prior approval of
the Chief Bridge Engineer.
3.11 EARTH PRESSURE: EH, ES, LS AND DD
3.11.1 General
Both the vertical and horizontal components of an
inclined lateral earth pressure shall be considered for
application of load and load factors.
3.11.3 Presence of Water
C3.11.3
Walls along a stream or river shall be designed for a
minimum differential water pressure due to a 1000 mm {3'0"} head of water in the backfill soil above the weephole
inverts.
Evaluation of water pressures and seepage forces is
critical in the design of retaining walls because water
pressures and seepage forces are the most common causes
of retaining wall failure. Seepage forces and water
pressures affect the stability of retaining walls by:
Increasing the weight of soil behind the wall through
saturation, thereby increasing the driving soil pressure
Decreasing the effective weight of soil in front of the
wall through upward seepage forces, thereby reducing
the resisting soil pressure
Decreasing the effective stress (normal force) on the
wall foundation due to wall weight through uplift,
thereby reducing sliding resistance and resistance to
overturning.
3.11.5 Earth Pressure: EH
C3.11.5.2 AT-REST PRESSURE COEFFICIENT, ko
At-rest earth pressures are usually limited to bridge
abutments to which superstructures are fixed prior to
backfilling (e.g., framed bridges) or to cantilevered walls
where the heel is restrained and the base/stem connection
prevents rotation of the stem.
B.3 - 37
SPECIFICATIONS
September 2007
COMMENTARY
C3.11.5.3 ACTIVE PRESSURE COEFFICIENT, ka
The differences between the Coulomb Theory currently
specified, and the Rankine Theory specified in the past is
illustrated in Figure C1. The Rankine theory is the basis of
the equivalent fluid method of A3.11.5.5 and the design
procedures for mechanically stabilized earth walls.
Gravity and semi-gravity walls usually deflect a
sufficient amount during backfilling to develop an active
state of stress in the retained soil. This also is true of
cantilevered and counterfort walls unless the heel is tied
down or otherwise restrained and the base/stem connection
prevents sufficient rotation of the stem to develop an active
state of stress in the soil.
Wall movements cause the development of friction
between the wall and the soil in contact with the wall. This
resulting frictional force has the effect of inclining the earth
pressure resultant on the wall, whereas the resultant would
be normal to the wall in the case of no friction. The angle of
inclination of the earth pressure resultant with respect to a
line normal to the wall is called the angle of wall
friction (δ).
3.11.5.4 PASSIVE PRESSURE COEFFICIENT, kp
The following shall replace Figures A3.11.5.4-1 and
A3.11.5.4-2.
B.3 - 38
SPECIFICATIONS
September 2007
COMMENTARY
Figure 3.11.5.4-1 - Computational Procedures for Passive Earth Pressure for Sloping Wall with Horizontal Backfill
B.3 - 39
September 2007
SPECIFICATIONS
COMMENTARY
Figure 3.11.5.4-2 - Computational Procedures for Passive Earth Pressure for Vertical Wall With Sloping Backfill
3.11.5.5
EQUIVALENT-FLUID
ESTIMATING EARTH PRESSURES
METHOD
OF
Cohesionless soils with a maximum fines content of 5%
by weight shall be used for backfill. This criteria can be met
by backfilling with AASHTO No. 57 or the Department's
open graded subbase (OGS) in conformance with
Publication 408, Section 703.
For yielding walls backfilled with these materials, the
design earth pressure at any depth shall be defined as
B.3 - 40
C3.11.5.5
In the fifth paragraph of AC3.11.5.5, remove the
reference to Figure AC3.11.5.3-1.
Soils with more than 5% fines shall be avoided as
backfill because of their low permeability and potential frost
susceptibility.
For design, the Department's open graded subbase
(OGS) shall have the following assumed properties:
SPECIFICATIONS
September 2007
COMMENTARY
increasing at a rate of 5.5 x 10-6 MPa/mm {0.035 ksf/ft},
plus the live load surcharge from A3.11.6.2 and D3.11.6.2.
For unyielding walls, restrained abutments (e.g.,
backfilled after superstructure erection), at-rest earth
pressure, increasing at 8.0 x 10-6 MPa/mm {0.05 ksf/ft},
plus the live load surcharge from A3.11.6.2 and D3.11.6.2,
shall be used.
The following shall supplement A3.11.5.5 for the
design of box culverts.
For box culverts, equivalent fluid density shall be taken
as specified in Table 2.
Table 3.11.5.5-2 - Equivalent Fluid Densities for Box
Culverts
Metric Units
U.S. Customary Units
Level
Backfil
l
(kg/m3
)
Backfill
with
β=25º
(kg/m3)
Level
Backfill
(kcf)
Backfill
with β=25º
(kcf)
Minimum
720
880
0.045
0.055
Maximum
1120
1280
0.070
0.080
These equivalent fluid densities along with the
appropriate maximum and minimum load factors shall be
selected to produce the extreme force effects.
moist density = 1920 kg/m3 {0.120 kcf}
saturated density = 2160 kg/m3 {0.135 kcf}
angle of internal friction = 30
The following shall supplement AC3.11.5.5 for the
design of box culverts.
Two soil types were selected for design to reflect
potential lateral at-rest earth pressures for box culverts,
considering construction practice and soil variability in
Pennsylvania. The engineered backfill required for a
distance of only 300 mm {1 ft.} from the face of the culvert
wall is not sufficient to reduce lateral earth pressures to
levels that would be expected for abutments and retaining
walls for which more detailed backfill requirements are
specified.
Lateral earth pressures resulting from the
factored load combinations, specified in this article, A3.4.1
and D3.4.1 compare closely with past DM-4 practice.
Although the equivalent fluid weights given in Table 1
correspond to those for "Dense Sand or Gravel" and
"Compacted Lean Clay", backfill material shall be in
conformance with the requirements given in Publication 408
and the contract documents. Equivalent fluid weights
specified herein are for design only. Values of the
equivalent fluid pressure for a sloping backfill are provided
for the rare case in which the culvert is parallel to the
roadway. In such a case, consideration should be given to
sliding as a result of the imbalance of lateral loads.
3.11.5.6 EARTH PRESSURE FOR NONGRAVITY
CANTILEVER WALLS
C3.11.5.6
For permanent walls, the simplified earth pressure
distributions shown in Figures 1 and 2, or other suitable
earth pressure distributions, may be used. If walls will
support or are supported by cohesive soils for temporary
applications, walls may be designed based on total stress
methods of analysis and undrained shear strength
parameters. For this latter case, the simplified earth
pressure distributions shown in Figures 3 and 4, or other
approved earth pressure distributions, may be used with the
following restrictions:
Nongravity cantilevered walls temporarily supporting
or supported by cohesive soils are subject to excessive
lateral deformation if the undrained soil shear strength is
low compared to the shear stresses. Therefore, use of these
walls should be limited to soils of adequate strength as
represented by the stability number (N = 10 -9γgH/c) {U.S.
Customary Units: N = γH/c}.
Base movements in the soil in front of a wall become
significant for values of N of about 3 to 4, and a base failure
can occur when N exceeds about 5 to 6, Terzaghi and Peck
(1967).
The ratio of overburden pressure to undrained shear
strength (i.e., stability number N = 10-9γgH/c {U.S.
Customary Units: N = γH/c}) must be < 3.
The active earth pressure shall not be less than 0.25
times the effective overburden pressure at any depth, or
5.5 x 10-6 MPa/mm {0.035 ksf/ft} of wall height,
whichever is greater.
For temporary walls with discrete vertical elements
B.3 - 41
SPECIFICATIONS
September 2007
COMMENTARY
embedded in granular soil or rock, Figure 1 may be used to
determine passive resistance and Figure 3 may be used to
determine the active earth pressure due to the retained soil.
B.3 - 42
SPECIFICATIONS
September 2007
COMMENTARY
Figure 3.11.5.6-1 - Simplified Earth Pressure Distributions for Permanent Nongravity Cantilevered Walls with Discrete
Vertical Wall Elements
B.3 - 43
SPECIFICATIONS
September 2007
COMMENTARY
Figure 3.11.5.6-2 - Simplified Earth Pressure Distributions for Permanent Nongravity Cantilevered Walls with Continuous
Vertical Wall Elements modified after Teng (1962)
B.3 - 44
SPECIFICATIONS
September 2007
COMMENTARY
Figure 3.11.5.6-3 - Simplified Earth Pressure Distributions for Temporary Nongravity Cantilevered Walls with Discrete
Vertical Wall Elements
B.3 - 45
SPECIFICATIONS
September 2007
COMMENTARY
Figure 3.11.5.6-4 - Simplified Earth Pressure Distributions for Temporary Nongravity Cantilevered Walls with Continuous
Vertical Wall Elements modified after Teng (1962)
B.3 - 46
September 2007
SPECIFICATIONS
COMMENTARY
Where discrete vertical wall elements are used for
support, the width of each vertical element shall be assumed
to equal the width of the flange or diameter of the element
for driven sections and the diameter of the concrete-filled
hole for sections encased in concrete.
In Figures 1 and 3, the width of discrete vertical wall
elements effective in mobilizing the passive resistance of the
soil is based on a method of analysis by Broms (1964a and
1964b) for single vertical piles embedded in cohesive or
cohesionless soil and assumes a vertical element. The
effective width for passive resistance of three times the
element width (b) is due to the arching action in soil and
side shear on resisting rock wedges. The maximum width
of 3b can be used when material in which the vertical
element is embedded is intact. This width shall be reduced
if planes or zones of weakness would prevent mobilization
of resistance through this entire width. If the element is
embedded in soft clay having a stability number less than 3,
soil arching will not occur and the actual width shall be used
as the effective width for passive resistance. Where a
vertical element is embedded in rock (Figure 1b), the
passive resistance of the rock is assumed to develop through
the shear failure of a rock wedge equal in width to the
vertical element (b) and defined by a plane extending
upward from the base of the element at an angle of 45 .
The magnitude and location of resultant loads and
resisting forces for permanent walls with discrete vertical
elements embedded in soil and rock for lateral support may
be determined using the earth pressure distributions
presented in Figures 1 and 3.
The procedure for
determining the resultant passive resistance of a vertical
element embedded in soil assumes the net passive resistance
is mobilized across a maximum of three times the element
width or diameter (reduced, if necessary, to account for soft
clay, or discontinuities in the embedded depth of soil or
rock) and that the active pressure below the facing elements
acts only on the actual vertical element width. For the
embedded portion of a wall with discrete vertical elements,
the net passive resistance shall not be taken greater than that
for continuous embedded vertical elements as determined
using Figure 2 for permanent walls and Figure 4 for
temporary walls.
The magnitude and location of resultant loads and
resisting forces for permanent walls with continuous vertical
elements may be determined using the earth pressure
distributions presented in Figure 2 for permanent walls and
Figure 4 for temporary walls.
Some portion of the embedded depth below finished
grade, noted as β’ in Figures 1-3, (usually 900 mm {3 ft.}
for an element in soil, and 300 mm {1 ft.} for an element in
rock) is ineffective in providing passive lateral support.
In developing the design lateral pressure, the lateral
pressure due to water, live load surcharge, permanent point
and line surcharge loads, backfill compaction, or other types
of surcharge loads shall be added to the lateral earth
pressure.
3.11.5.7
APPARENT
ANCHORED WALLS
EARTH
PRESSURES
FOR
B.3 - 47
The upper 600 to 900 mm {2 to 3 ft.} of the discrete
embedded vertical element in soil, or 300 mm {1 ft.} in
rock, is typically assumed ineffective in mobilizing passive
resistance to account for the effects of freezing and thawing,
weathering or other shallow ground disturbance (e.g., utility
excavations or pavement replacement in front of the wall).
C3.11.5.7
SPECIFICATIONS
COMMENTARY
The following shall replace A3.11.5.7:
For anchored walls constructed from the top down, the
earth pressure may be estimated in accordance with Articles
D3.11.5.7.1 or D3.11.5.7.2.
3.11.5.7.1 Cohesionless Soils
Metric Units:
0.65k a ' S gH 2 (10
9
)
(3.11.5.7.1-3)
Pa
0.65k a ' S H 2
where:
Pa =
total earth pressure load per unit length of wall
(N/mm of wall) {kip/ft of wall}
ka
active earth pressure coefficient (dim)
tan2(45- f/2) for β=0
use Equation A3.11.5.3-1 for β 0
=
=
γ’S =
AC3.11.5.7.
The earth pressure diagrams used herein are primarily
intended for use in homogeneous soils. They should not be
used indiscriminately in stratified or relatively nonhomogeneous soil layers; engineering judgment must be
used in these cases.
When anchors, especially those near the top of the wall,
are tensioned to loads in excess of those estimated using the
apparent pressure diagrams, it is possible that the wall could
be displaced back into the soil mass, resulting in undesirable
deflections or a passive failure of the retained soil. It is
important to remember that anchored walls are flexible and
that they derive their satisfactory performance from a match
between the soil pressure and the wall-anchor loads.
C3.11.5.7.1P
The following shall supplement A3.11.5.7.1:
The apparent earth pressure distribution for temporary
and permanent anchored walls constructed from the top
down and supporting cohesionless soil may be determined
using Figures 1(a) and 1(b). Water pressures and surcharge
pressures, if applicable, should be added explicitly to the
diagrams to evaluate the total lateral load acting on the wall.
Determine geostatic water pressure on the wall using the
maximum expected water table differential between
excavation interior and exterior, based on borings or other
information.
In both Figures 1(a) and 1(b), calculate the maximum
pressure ordinate p as indicated in the figures.
The earth pressure total load Pa per unit length of wall
may then be calculated from the area of the apparent earth
pressure distribution diagram as:
Pa
September 2007
effective unit weight of soil (kg/m3) {kcf}
B.3 - 48
Anchored walls are typically constructed with freedraining material placed immediately behind the lagging,
and therefore geostatic water pressure on the wall would not
be of concern. However, there may be conditions of a
permanent water table behind the wall where geostatic water
pressure needs to be considered.
SPECIFICATIONS
H
=
COMMENTARY
total excavation depth (mm) {ft.}
H1 =
distance from ground surface to uppermost ground
anchor (mm) {ft.}
Hn+1 =
distance from base of excavation to lowermost
ground anchor (mm) {ft.}
Thi =
horizontal load in anchor I (N/mm of wall) {kip/ft
of wall}
R
reaction force to be resisted by subgrade (i.e.,
below base of excavation) (N/mm of wall) {kip/ft
of wall}
=
September 2007
B.3 - 49
SPECIFICATIONS
September 2007
COMMENTARY
Figure 3.11.5.7.1-1(a) – Apparent Earth Pressure
Distribution for Anchored Walls Constructed from the top
down in Cohesionless Soils
Figure 3.11.5.7.1-1(b) – Apparent Earth Pressure
Distribution for Anchored Walls Constructed from the top
down in Cohesionless Soils
B.3 - 50
September 2007
SPECIFICATIONS
COMMENTARY
3.11.5.7.2 Cohesive Soils
C3.11.5.7.2P
The following shall replace A3.11.5.7.2:
The apparent earth pressure distribution for cohesive
soils is related to the stability number, Ns, which is defined
as:
Cohesive soils with a stability number NS 4 are to be
considered to be stiff to hard in consistency. Cohesive soils
with a stability number NS > 4 are to be considered very soft
to medium-stiff in consistency
Metric Units:
S
NS
gH (10 9 )
Su
(3.11.5.7.2-1)
SH
Su
Ns
where:
NS =
stability number (dim)
γS
=
total unit weight of soil (kg/m3) {kcf}
H
=
total excavation depth (mm) {ft.}
Su =
average undrained shear strength of soil (MPa)
{ksf}
Use the undrained shear strength of the soil through
which the excavation extends.
3.11.5.7.2a Stiff to Hard, Including Fissured Cohesive Soils
The following shall replace A3.11.5.7.2a:
anchored walls constructed from the top down and
supporting stiff to hard cohesive soils (NS 4) including
fissured clays, where temporary conditions are of a
controlled short duration and for which there is no available
free water, may be determined using Figures 3.11.5.7.1-1(a)
and 3.11.5.7.1-1(b). The identified terms remain the same,
except that the maximum pressure ordinate of the diagram p
shall be calculated as:
Metric Units:
p
0.2 s gH 10
9
to 0.4 s gH 10
9
(3.11.5.7.2a-1)
p
0.2 s H to 0.4 s H
B.3 - 51
C3.11.5.7.2a
The following shall supplement AC3.11.5.7.2:
SPECIFICATIONS
September 2007
COMMENTARY
where:
p
=
maximum pressure ordinate (MPa) {ksf}
γS
=
total unit weight of soil (kg/m3) {kcf}
H
=
total depth of excavation (mm) {ft.}
Surcharge pressures, if applicable, should be added
explicitly to the diagrams to evaluate the total lateral load
acting on the wall.
For other temporary conditions and for permanent
conditions, calculate the earth pressure resultant using the
maximum pressure ordinate obtained by either Equation 1 or
Equation 3.11.5.7.1-1 with a value of ka based on the
drained friction angle of the clay. For any case, surcharge
pressures, if applicable, should be added explicitly to the
diagrams to evaluate the total lateral load acting on the wall.
For conditions where there is available free water, determine
geostatic water pressure on the wall using the maximum
expected water table differential between excavation
interior and exterior, based on borings or other information.
For permanent walls, the distribution (permanent or
temporary) resulting in the maximum total force shall be
used for design.
Alternatively, in fissured clays the apparent earth
pressure diagram may be based upon previous successful
experience with excavations constructed in similar soils.
This is because earth pressures in these soils are most
influenced by degree of fissuring or jointing in the clay and
the potential reduction in strength with time, not necessarily
the shear strength of the intact clay.
3.11.5.7.2b Very Soft to Medium-Stiff Cohesive Soils
The following shall replace A3.11.5.7.2b:
and permanent anchored walls constructed from the top
down and supporting soft to medium-stiff cohesive soils
may be determined using Figure 1. Soft to medium-stiff
cohesive soils are those with a stability number N S > 4.
B.3 - 52
There may be conditions of a permanent water table
behind the wall where geostatic water pressure needs to be
considered.
SPECIFICATIONS
COMMENTARY
Figure 3.11.5.7.2b-1 - Apparent Earth Pressure Distribution
for Anchored Walls Constructed from the top down in Soft
to Medium-Stiff Cohesive Soils
Calculate the maximum pressure ordinate of the
diagram p as:
Metric Units:
p
1.0 k a
S
g H (10 9 )
(3.11.5.7.2b-1)
p
1.0 k a
S
H
where:
ka
=
0.22 for 4
NS < 5.14
or
Metric Units:
ka
4S u
1 m
S
gH (10 9 )
(3.11.5.7.2b-2)
ka
1 m
for NS
September 2007
4S u
SH
5.14, and using m = 0.4
B.3 - 53
September 2007
SPECIFICATIONS
COMMENTARY
Additionally, if NS 6 and the excavation is underlain
by soft clay, calculate ka by Equation 2 and Equation 3
below, and use the larger of the 2 ka values in Equation 1 to
calculate the maximum pressure ordinate.
Metric Units:
ka
4S u
1
9
)
S gH (10
2 2d
1
H
H
a
H
S ub
9
)
S gH (10
b
ka
4S u
SH
1
2 2d
1
H
H
a
H
S ub
b
SH
where
a
H
1
H
(2 x ) 2d (dim)
b
514
.
2S u H
2 S ub d (dim)
Su =
undrained shear strength of retained soil (MPa)
{ksf}
Sub =
undrained shear strength of soil providing bearing
resistance (MPa) {ksf}
d
=
depth of the potential base failure surface below the
base of excavation (mm) {ft.}
γS
=
total unit weight of retained soil (kg/m3) {kcf}
ΔH =
depth of unloading at ground surface, if any (mm)
{ft.}
X
length of unloading at top of anchored wall
excavation, if any (mm) {ft.}
=
The value of d is taken as the thickness of soft to
medium-stiff cohesive soil below the excavation base up to
a maximum value of Bc
width.
2 , where Bc is the excavation
In any case, surcharge pressures, if applicable, should
be added explicitly to the diagrams to evaluate the total
lateral load acting on the wall. For conditions where there is
available free water, determine geostatic water pressure on
the wall using the maximum expected water table
differential between excavation interior and exterior, based
B.3 - 54
There may be conditions of a permanent water table
behind the wall where geostatic water pressure needs to be
considered.
September 2007
SPECIFICATIONS
COMMENTARY
on borings or other information.
3.11.5.8 EARTH PRESSURES FOR MECHANICALLY
STABILIZED EARTH WALLS
3.11.5.8.1 General
The following shall replace the definition of ka in
A3.11.5.8.1:
ka
=
active earth pressure coefficient specified herein
The following shall supplement A3.11.5.8.1:
Lateral earth pressure coefficients for MSE walls may
be determined as follows:
for a horizontal or sloping backfill surface, as shown in
Figures A3.11.5.8.1-1 and A3.11.5.8.1-2, active earth
pressure coefficient, ka, in determining safety against
soil failure may be taken as:
ka
β
f
cos
cos 2
cos 2
cos
2
2
cos
cos
cos
f
(3.11.5.8.1-2)
f
=
slope of backfill behind wall (DEG)
=
internal friction angle of backfill soil (DEG)
for a broken back backfill surface, the active earth
pressure coefficient, ka, for evaluation of safety against
soil failure may be taken as:
Error! Objects cannot be created from editing field
(3.11.5.8.1-3)
codes.
where:
B
=
notional slope of backfill behind wall as shown in
Figure A3.11.5.8.1-3 (DEG)
f
=
internal friction angle (DEG)
active earth pressure coefficient, ka,
determining safety against structural failure:
ka
tan 2 45o
f
2
for
(3.11.5.8.1-4)
3.11.6 Surcharge Loads: ES and LS
3.11.6.4 LIVE LOAD SURCHARGE: LS
C3.11.6.4
B.3 - 55
September 2007
SPECIFICATIONS
COMMENTARY
The following shall replace Table A3.11.6.4-1.
Table 3.11.6.4-1 - Equivalent Height of Soil for Vehicular
Loading – Abutment
Metric Units
Wall Height
(mm)
Heq
(mm)
Wall Height
(ft)
Heg
(ft)
≤ 1500
1200
≤ 5.0
4.0
≥ 3000
900
≥ 10.0
3.0
The minimum design surcharge values for abutments in
Table 1 are intended to account for normal traffic live loads
and do not address the effects of backfill compaction. Refer
to A3.11.2 to determine the effects of backfill compaction.
For retaining walls, use Table 2.
Table 3.11.6.4-2 - Equivalent Height of Soil (heq) for
Vehicular Loading - Retaining Walls
Metric Units
Wall
Height
(mm)
Distance from
back face of
wall to the
wheel line
0.0mm
300mm
≤ 1500
1500
900
3000
1050
≥ 4000
900
Wall
Height
(ft)
Distance from
back face of
wall to the
wheel line
0.0 ft
1 ft
≤ 5.0
5.0
3.0
900
10.0
3.5
3.0
900
≥ 13.0
3.0
3.0
Delete the third paragraph of AC3.11.6.4.
In the development of this specification, the
Department had a comparison made between their past
abutment and retaining wall service load design method and
the LRFD method. With minor modifications contained in
this specification, the LRFD method gave similar results to
the Department's past design method with one exception.
For walls less than 1500 mm {5 ft.} in height on poor soils,
the LRFD method may require base width significantly
larger than past designs. Since the Department has not
experienced problems with short headwalls for pipe
culverts, the Standard Drawings may be used for headwalls
for pipe culverts.
In Table D.3.11.6.4-2, the distance from back face of
wall to edge of traveled way of 0.0 mm {0 ft} corresponds
to placement of a point wheel load 600 mm {2 ft} from the
back face of the wall. For the case of the uniformly
distributed lane load, the 0.0 mm {0 ft} distance
corresponds to the edge of the 3000 mm {10 ft} wide traffic
lane.
For box culverts, use 900 mm {3.0 ft} where live load
effects are considered.
3.12 FORCE EFFECTS DUE TO SUPERIMPOSED
DEFORMATIONS: TU, TG, SH, CR, SE
3.12.2 Uniform Temperature
3.12.2.1.1 TEMPERATURE RANGES
C3.12.2.1
Provision shall be made for forces and movements
resulting from variations in temperature. The range of
temperature with respect to the normal erection temperature
of 20 C {68 F} shall be as given in Table 1.
The inclusion of an additional temperature fall of 32 C
is based on a Departmental study conducted in District 3-0.
It was determined that the fixity at the connections of
continuous spans produces a frame-type action that induces
additional forces.
B.3 - 56
September 2007
SPECIFICATIONS
COMMENTARY
Table 3.12.2.1.1-1 – Procedure A Temperature Ranges
Metric Units
Material
Temperature Rise
Temperature Fall
Temperature Rise
Temperature Fall
Steel or Aluminum Structures
23 C
43 C
42 F
78 F
Concrete Structures
18 C
32 C
32 F
58 F
4 C
8 C
7 F
14 F
Neoprene
Other
Neoprene
Other
Bearings (Prestressed Concrete
Structures) Temp. Range
45 C
64 C
80 F
116 F
Bearings (Steel or Aluminum
Structures) Temp. Range
56 C
86 C
100 F
156 F
Wood Structures
For the design of integral abutments, the temperature
range given for bearings shall be used.
3.12.3 Temperature Gradient
C3.12.3
The load factor for temperature gradient shall be taken
as zero for those bridges which can be analyzed by the
approximate methods given in A4.6.2 and D4.6.2, and are of
Type a, b (only precast P/S concrete box girders), e, f, g, h,
j, k and l as given in Table A4.6.2.2.1-1.
For Pennsylvania bridges other than those listed above,
the Zone 3 data shall be used as given in Table A3.12.3-1.
Pennsylvania has not experienced any temperature
gradient-related problems in their typical multi-girder
bridges. Therefore, as suggested in AC3.12.3, the
Department's experience with typical multi-girder bridges
has led them to exclude the temperature gradient load
condition for these types of bridges.
C3.12.7P
3.12.7P
Minimum Temperature Force for Fixed
Substructures
When neoprene bearings are used, the fixed
substructure unit(s) shall consider a thermal force equal to
the largest thermal force from the largest expansion bearing
substructure unit or utilize the results of an equilibrium
analysis, whichever is larger.
3.12.8P Temporary Support Settlement For Curved and
Skewed Bridges During Construction
When a temporary falsework is used, an analysis should
be performed to check its settlement effects on member
response during construction. As a minimum, the following
scenarios should be considered for the analysis:
Settlement of single and multiple temporary supports.
A minimum settlement of one thousandth of the span
length should be used.
B.3 - 57
This provision insures that fixed substructures are
designed for a minimal thermal force even if an equilibrium
analysis indicates no thermal forces are present. This is
similar to the forces applied to steel bearings considering
frozen bearings.
C3.13
The following shall replace AC3.13.
Low and high friction coefficients may be obtained
from standard textbooks. If so warranted and approved by
the Chief Bridge Engineer, the values may be determined by
physical tests, especially if the surfaces are expected to be
roughened in service.
When a force is transmitted from the superstructure to
the substructure through a sliding bearing, the force applied
to the substructure is considered a frictional force.
However, forces transmitted, via a non-sliding bearing such
as an elastomeric bearing, are factored by the appropriate
load factor for the driving effect, such as TU.
SPECIFICATIONS
September 2007
COMMENTARY
3.13 FRICTION FORCES: FR
The following shall supplement A3.13.
Friction force acts parallel to the direction of movement
and is assumed to act at the bearing elevation at each
expansion bearing, with due consideration given to the
reactions that must develop at the fixed bearings to satisfy
equilibrium. See A14.6.3.1 for horizontal forces.
Consideration of frozen expansion bearings and
variation of friction is provided assuming the largest pier or
abutment DL reaction times the applicable friction
coefficient acts at the fixed pier or utilize the results of an
equilibrium analysis, whichever is larger.
3.14 VESSEL COLLISION: CV
3.14.1 General
C3.14.1
The vessel collision provisions provided in A3.14 and
D3.14 shall only be used in the substructure design of
bridges which cross a navigable waterway. The Department
defines a navigable waterway as those waterways which:
For the vast majority of bridges over waterways in
Pennsylvania, the vessel collision provisions will not be
applicable.
The vessel collision provisions will most likely be
applicable for bridges over the following waterways:
presently support commercial barge and/or ship traffic,
and
have supported commercial barge and/or ship traffic
within the past 20 years.
lower portions of Delaware River
lower portions of Schuylkill River
lower portions of Allegheny River
There is some reason to believe that the waterway will
support commercial barge and/or ship traffic in the
future.
lower portions of Monongahela River
Ohio River
3.14.2 Owner's Responsibility
When the vessel collision provisions are applicable
according to D3.14.1, the designer must submit at the Type,
Size and Location stage the following information for
review by the Department:
vessel traffic density in the waterway
design velocity of vessels for the bridge
suggested degree of damage that the bridge
components, including protective systems are allowed
to sustain
C3.14.15 Protection of Substructures
B.3 - 58
September 2007
SPECIFICATIONS
COMMENTARY
Any testing for protection systems of substructures
must be approved by the Chief Bridge Engineer.
3.15P FORCE TRANSFER TO SUBSTRUCTURE
3.15.1P Longitudinal Force
3.15.1.2P FORCE TRANSFER TO SUBSTRUCTURE
Longitudinal forces, except friction (see D3.13), shall
be carried only by fixed bearings.
3.15.1.3P
EFFECTIVE
SUPERSTRUCTURE FORCES
LENGTH
FOR
Longitudinal forces transmitted to the substructure from
the superstructure shall be calculated using the center-tocenter bearing length of superstructure restrained by fixed
bearings. In the case of consecutively fixed piers, forces to
the substructure shall be determined with due consideration
to the relative stiffness of the piers.
3.15.1.4P FORCE RESOLUTION TO SUBSTRUCTURE
Longitudinal forces from the superstructure shall be
directly applied at the bearings and shall be resolved in the
directions perpendicular and parallel to the substructure, as
shown in Figure 1. For frame analysis of the substructure,
an equivalent parallel component shall be used, as shown in
Figure 2.
For structures on a sloping grade with an inclined
bearing plate, the reaction component parallel to the grade
(longitudinal force) shall be considered.
B.3 - 59
September 2007
SPECIFICATIONS
COMMENTARY
Figure 3.15.1.4P-1 - Force Resolution to Substructure
Figure 3.15.1.4P-2 - Equivalent Force for Frame Analysis of
Substructure
3.15.2P Transverse Force
3.15.2.1P FORCE TRANSFER TO SUBSTRUCTURE
The transverse forces applied to the superstructure must
be resisted by the bearings.
3.15.2.2P
EFFECTIVE
LENGTHS
FOR
B.3 - 60
SPECIFICATIONS
COMMENTARY
SUPERSTRUCTURE FORCES
Unless a more rational method of analysis is used,
transverse forces acting on a superstructure shall be
transmitted to the bearings using the following span lengths:
Continuous
Spans
Simple Spans
Piers
Average of the two
adjacent spans
Abutments
One-half of the end
span
One-half of the span
3.15.2.3P FORCE RESOLUTION TO SUBSTRUCTURE
Transverse forces from the superstructure shall be
resolved in the directions perpendicular and parallel to the
substructure, as shown in Figure D3.15.1.4P-1.
3.15.2.4P
REACTIONS
September 2007
DETERMINATION
OF
BEARING
The effect of the transverse force applied at the
elevation specified for that force shall be taken into account
in determining the vertical reactions at the bearings (see
Figure 1).
Figure 3.15.2.4P-1 - Bearing Reactions
B.3 - 61
September 2007
REFERENCES
American Concrete Pipe Association (ACPA), Concrete Pipe Handbook, Vienna, VA, 435 p., 1981
American Society of Civil Engineers (ASCE), Design and Construction of Sanitary and Storm Sewer, prepared by a joint
committee of the ASCE and the Water Pollution Control Federation (WPCF), ASCE - Manuals and Reports of Engineering
Practice - No. 37 (WPCF Manual of Practice No. 9), 1969, 350 p.
Broms, B. B., "Lateral Resistance of Piles in Cohesive Soils", Journal of Soil Mechanics and Foundation Engineering
Division, ASCE, Vol. 90, No. SM2, pp. 27-64, 1964a.
Broms, B. B., "Lateral Resistance of Piles in Cohesive Soils", Journal of Soil Mechanics and Foundation Engineering
Division, ASCE, Vol. 90, No. SM3, pp. 123-156, 1964b.
Teng, W. C., "Foundation Design", Prentice-Hall, Englewood Cliffs, New Jersey, 466 pp., 1962
B.3 - 62
DESIGN MANUAL
PART 4
VOLUME 1
SECTION 4 - STRUCTURAL ANALYSIS AND EVALUATION
SCOPE ........................................................................................................................................................................ B.4 - 1
DEFINITIONS ........................................................................................................................................................... B.4 - 1
ACCEPTABLE METHODS OF STRUCTURAL ANALYSIS ............................................................................. B.4 - 1
MATHEMATICAL MODELING ........................................................................................................................... B.4 - 2
4.5.1 General ............................................................................................................................................................. B.4 - 2
4.5.2 Structural Material Behavior ......................................................................................................................... B.4 - 2
4.5.2.2 ELASTIC BEHAVIOR ........................................................................................................................... B.4 - 2
4.5.2.3 INELASTIC BEHAVIOR ....................................................................................................................... B.4 - 3
4.6 STATIC ANALYSIS ................................................................................................................................................. B.4 - 3
4.6.1 Influence of Plan Geometry ............................................................................................................................ B.4 - 3
4.6.1.2 STRUCTURES CURVED IN PLAN ...................................................................................................... B.4 - 3
4.6.1.2.1 General .......................................................................................................................................... B.4 - 3
4.6.2 Approximate Methods of Analysis ................................................................................................................. B.4 - 3
4.6.2.1 DECKS .................................................................................................................................................... B.4 - 3
4.6.2.1.3 Width of Equivalent Interior Strips ............................................................................................... B.4 - 3
C4.6.2.1.6 Calculation of Force Effects ............................................................................................................................. B.4 - 3
4.6.2.1.8 Live Load Distribution on Fully Filled and Partially Filled Grids ................................................ B.4 - 3
4.6.2.1.9 Inelastic Analysis .......................................................................................................................... B.4 - 4
4.6.2.2 BEAM-SLAB BRIDGES ........................................................................................................................ B.4 - 4
4.6.2.2.1 Application .................................................................................................................................... B.4 - 4
4.6.2.2.2 Distribution Factor Method for Moment and Shear ...................................................................... B.4 - 8
4.6.2.2.2a Interior Beams with Wood Decks ......................................................................................... B.4 - 8
4.6.2.2.2b Interior Beams with Concrete Decks .................................................................................... B.4 - 8
4.6.2.2.2e Skewed Bridges ................................................................................................................... B.4 - 13
4.6.2.2.3 Distribution Factor Method for Shear ......................................................................................... B.4 - 13
4.6.2.2.3a Interior Beams .................................................................................................................... B.4 - 13
4.6.2.3 EQUIVALENT STRIP WIDTHS FOR SLAB-TYPE BRIDGES ........................................................ B.4 - 19
4.6.2.5 EFFECTIVE LENGTH FACTOR, K .................................................................................................... B.4 - 19
4.6.2.6.1 General ........................................................................................................................................ B.4 - 20
4.6.2.10P GIRDER - FLOORBEAM - STRINGER BRIDGES ....................................................................... B.4 - 21
4.6.2.10.1P Girder Live Load Distribution Factors .................................................................................... B.4 - 21
4.6.2.10.2P Stringer Live Load Distribution Factors.................................................................................. B.4 - 21
4.6.2.10.3P Floorbeam Live Load Distribution Factors ............................................................................. B.4 - 21
4.6.2.10.3aP Floorbeams with the Top Flange not Directly Supporting the Deck ............................. B.4 - 21
4.6.2.10.3bP Floorbeams with the Top Flange Directly Supporting the Deck ................................... B.4 - 21
4.6.2.11P DISTRIBUTION OF LOAD FROM THE SUPERSTRUCTURE TO THE SUBSTRUCTURE .... B.4 - 21
4.6.2.12P EQUIVALENT STRIP WIDTHS FOR BOX CULVERTS .............................................................. B.4 - 22
4.6.2.12.1P General ..................................................................................................................................... B.4 - 22
4.6.2.12.2P Case 1: Traffic Travels Parallel to Span ................................................................................... B.4 - 22
4.6.2.12.3P Case 2: Traffic Travels Perpendicular to Span ......................................................................... B.4 - 22
4.6.2.12.4P Precast Box Culverts ............................................................................................................... B.4 - 23
4.6.3 Refined Methods of Analysis ........................................................................................................................ B.4 - 24
4.6.3.1 GENERAL ............................................................................................................................................ B.4 - 24
4.6.3.2 DECKS .................................................................................................................................................. B.4 - 24
4.6.3.2.3 Orthotropic Plate Model .............................................................................................................. B.4 - 24
4.6.3.3 BEAM-SLAB BRIDGES ...................................................................................................................... B.4 - 24
4.1
4.2
4.4
4.5
B.4 - i
4.6.4 Redistribution of Negative Moments in Continuous Beam Bridges .......................................................... B.4 - 36
4.6.4.1 GENERAL ............................................................................................................................................ B.4 - 36
4.6.4.2 REFINED METHOD ............................................................................................................................ B.4 - 36
4.6.4.3 APPROXIMATE PROCEDURE .......................................................................................................... B.4 - 36
4.7 DYNAMIC ANALYSIS .......................................................................................................................................... B.4 - 36
4.7.1 Basic Requirements of Structural Dynamics............................................................................................... B.4 - 36
C4.7.1.4 DAMPING .......................................................................................................................................... B.4 - 36
4.7.2 Elastic Dynamic Responses ........................................................................................................................... B.4 - 36
4.7.2.2 WIND-INDUCED VIBRATION .......................................................................................................... B.4 - 36
4.7.2.2.1 Wind Velocities ........................................................................................................................... B.4 - 36
4.7.4 Analysis for Earthquake Loads .................................................................................................................... B.4 - 36
4.7.4.3 MULTI-SPAN BRIDGES ..................................................................................................................... B.4 - 36
4.7.4.3.1 Selection of Method .................................................................................................................... B.4 - 36
4.7.4.3.5P Determination of Elastic Forces and Displacements ................................................................. B.4 - 37
4.7.4.4 MINIMUM DISPLACEMENT REQUIREMENTS ............................................................................. B.4 - 37
4.7.4.5P BASE ISOLATION DESIGN ............................................................................................................. B.4 - 38
4.7.4.5.1P General ...................................................................................................................................... B.4 - 38
4.7.4.5.2P Statically Equivalent Seismic Force and Coefficient ................................................................ B.4 - 43
4.7.4.5.3P Requirements for Elastic Force Determination ......................................................................... B.4 - 45
4.7.4.5.4P Design Displacement for Other Loads ...................................................................................... B.4 - 46
4.7.4.5.5P Design Forces for Seismic Zone 1 ............................................................................................ B.4 - 46
4.7.4.5.6P Design Forces for Seismic Zone 2 ............................................................................................ B.4 - 46
4.7.4.5.7P Substructure Design Requirements ........................................................................................... B.4 - 46
4.7.4.5.7.1P Foundations and Abutments ........................................................................................... B.4 - 46
4.7.4.5.7.2P Columns, Footings and Connections .............................................................................. B.4 - 46
4.7.4.5.7.2aP Structural Steel .............................................................................................................. B.4 - 46
4.7.4.5.7.2bP Reinforced Concrete...................................................................................................... B.4 - 46
4.8 ANALYSIS BY PHYSICAL MODELS ................................................................................................................. B.4 - 47
4.8.2 Bridge Testing ................................................................................................................................................ B.4 - 47
B.4 - ii
DM-4, Section 4 – Structural Analysis and Evaluation
September 2007
SPECIFICATIONS
COMMENTARY
4.1 SCOPE
C4.1
The following shall replace the last paragraph of A4.1.
Bridge structures shall be analyzed elastically, except as
noted herein. An inelastic analysis of bridge structures may
only be used for Extreme Event Limit States with the
approval of the Chief Bridge Engineer.
Delete the last sentence of the last paragraph of AC4.1.
4.2 DEFINITIONS
Automatic Mesh Generator - Program or subprogram which creates the layout (arrangement of nodes and elements) of a
model for the user if certain basic information is provided.
Influence Surface - Curved surface on which the ordinate is the value of the function (shear, moment, reaction, etc.) when a
unit load is placed at the ordinate for a member location (centerline of a girder, support, etc.).
Influence Surface Loader - Computer program or portion of a computer program which calculates and maximizes the value
of a function (shear, moment, reaction, deflection, etc.) by using the influence surface for that function.
Line Girder Analysis - Analysis of a bridge in which each girder is removed and analyzed as a single non-interacting
element.
Loading Algorithm - Methodology used by the influence surface loader to calculate the moments, shear, etc.
PEP - Project Engineering Panel of Applied Technology Council
Refined Analysis - Analysis according to A4.6.3 and D4.6.3.
Shear Lag - Nonuniform stress pattern due to ineffective transmission of shear.
Skew Angle - Angular measurement between the base line of the bridge and centerline of the pier; a 90 skew angle defining
a right bridge.
St. Venant Torsion - Uniform torsion resulting in no deformation of the cross-section.
Three-Dimensional Finite Element Analysis - Analysis in which a three-dimensional continuum is modeled as an
assemblage of discrete elements in three-dimensional space.
Warping Torsion - Nonuniform torsion resulting in warping of the cross-section.
4.4 ACCEPTABLE METHODS OF STRUCTURAL
ANALYSIS
The designer shall also follow the requirements in
PP1.4 in regards to computer programs.
Any computer program for the "3D or refined" analysis
of girder bridges which has not been reviewed by the
Department shall be submitted to, and approved by, the
Chief Bridge Engineer prior to its use. A sample bridge(s)
selected by the Department is to be modeled with the
program so that the Department can make comparisons
between its reviewed programs and the proposed program.
B.4 - 1
September 2007
SPECIFICATIONS
COMMENTARY
Computer programs for the analysis of girder bridges
approved for use (on LFD design projects) are listed in
Appendix J. Only the version of a program listed in
Appendix J has been tested and approved. If any changes
and/or modifications have been made to a program since its
approval date, then re-approval of the program is required.
The approval of these programs is subject to the following
conditions and limitations:
1.
While certain software packages provide design
optimization and/or code compliance checks, these
aspects were not included in the review and approval
process. Acceptance has been based solely upon the
review of generalized design forces (moments, shears,
reactions, etc.), as calculated by the software.
2.
Acceptance of a software package by the Department
does not affect the responsibility of the user for the
proper application of the software and interpretation of
its results. The acceptance of a software package does
not constitute an endorsement nor does it relieve the
vendor and the designer from their responsibility for
accurate, technically correct and sound engineering
results and services to the Department.
3.
The Department's acceptance does not constitute any
form of implied warranty, including warranty of
merchantability and fitness for a particular purpose.
The Commonwealth makes no warranty or
representation, either expressed or implied, with respect
to this software or accompanying documentation,
including their quality performance, merchantability, or
fitness for a particular purpose. In addition, the
Commonwealth will not be liable for any direct,
indirect, special, incidental, or consequential damages
arising out of the use, inability to use, or any defect in
the software or any accompanying documentation.
4.5 MATHEMATICAL MODELING
4.5.1 General
A4.5.1.
Barriers shall not be considered in the calculation of the
structural stiffness nor structural resistance of a structure.
Centerline distances shall be used in the analysis of
continuous frames, such as boxes, arches and pier bents.
4.5.2 Structural Material Behavior
4.5.2.2 ELASTIC BEHAVIOR
For simple and continuous spans, composite stiffness
B.4 - 2
September 2007
SPECIFICATIONS
COMMENTARY
shall be used if a concrete deck is used.
4.5.2.3 INELASTIC BEHAVIOR
The following shall replace the first sentence of the
second paragraph of A4.5.2.3.
The inelastic model shall be based either upon the
results of physical tests or upon a representation of load
deformation behavior which is validated by tests, but either
method must be approved by the Chief Bridge Engineer.
4.6 STATIC ANALYSIS
4.6.1 Influence of Plan Geometry
4.6.1.2 STRUCTURES CURVED IN PLAN
4.6.1.2.1 General
C4.6.1.2.1
Bridges which have kinked girders shall use the
provisions of A4.6.1.2.1 to determine if they are to be
considered curved.
For the design of horizontally curved steel girder
highway bridges, a load and resistance factor design is
required. A load factor design may also be permitted if
requested at TS&L stage. For a load factor design, use the
AASHTO, Guide Specification for Horizontally Curved
Highway Bridges, AASHTO, Standard Specifications for
Highway Bridges, and the 1993 Design Manual, Part 4. The
force effects (i.e., moments, shear, reacting, etc.) for the
curved bridge shall be determined using a refined method of
analysis. (Please note that the above-referenced LFD
documents are in U. S. Customary Units.)
The selected refined method of analysis for a structure
curved in plan must provide an accurate prediction of
behavior, both during construction and while in-service.
While the method of analysis that is selected is at the
discretion of the designer, a superstructure modeling
technique that represents the girder webs and concrete deck
using shell elements and other major superstructure
components using beam elements provides an acceptable
compromise between reduced computation times provided
by grillage analogy models and increased accuracy
provided by more sophisticated three-dimensional finite
element models.
4.6.2 Approximate Methods of Analysis
4.6.2.1 DECKS
4.6.2.1.3 Width of Equivalent Interior Strips
In Table A4.6.2.1.3-1 replace the entry for wood planks
as follows:
The width of a primary wood plank strip spanning
parallel to traffic shall be taken as 510 mm {20 in.}. The
width of a primary wood plank strip spanning perpendicular
to traffic shall be taken as the plank width, but not less than
250 mm {10 in.}.
C4.6.2.1.6 Calculation of Force Effects
Delete the first paragraph of AC4.6.2.1.6.
4.6.2.1.8 Live Load Distribution on Fully Filled and
Partially Filled Grids
Design in accordance with Tables per BD-604M.
B.4 - 3
September 2007
SPECIFICATIONS
COMMENTARY
The stiffness ratio, D, shall be taken as:
for fully filled grids with at least 40 mm {1 1/2 in.}
monolithic overfill ......................................................2.0
for all other fully filled grids ......................................2.5
for partially filled grids with at least 40 mm {1 1/2 in.}
monolithic overfill ......................................................8.0
for all other partially filled grid ................................10.0
When approved by the Chief Bridge Engineer, the
stiffness ratio, D, determined from test results may be used.
4.6.2.1.9 Inelastic Analysis
The following shall replace A4.6.2.1.9
The inelastic finite element analysis or yield line
analysis are not permitted unless specifically approved by
the Chief Bridge Engineer. If approved, this type of
analysis is to be only used for Extreme Event Limit State.
4.6.2.2 BEAM-SLAB BRIDGES
4.6.2.2.1 Application
C4.6.2.2.1
The following shall replace the third paragraph of
A4.6.2.2.1.
For any variables exceeding the range of applicability,
as specified in A4.6.2.2 and D4.6.2.2, the Chief Bridge
Engineer must approve the method for determining the
distribution factors.
The articles in this section which provide approximate
distribution factors are not applicable for bridges which are
considered curved as defined in A4.6.1.2.1 and D4.6.1.2.1.
For curved bridges, a refined method of analysis, as defined
in A4.6.3 and D4.6.3, is required.
Additional requirements for skewed structures must be
considered as follows:
The following shall replace the fifth sentence of the
twelth paragraph of AC4.6.2.2.1.
The use of transverse mild steel rods secured by nuts, or
similar unstressed dowels should not be considered
sufficient to achieve full transverse flexural continuity
unless demonstrated by test or experience and approved by
Apply the skew adjustment factors as given in
A4.6.2.2.3c and D4.6.2.2.3c on all skewed structures as
a minimum.
Steel structures with a skew angle less than 70 require
an additional check against uplift at the acute corners.
Concrete structures with a skew angle less than 45
require an additional check against uplift at the acute
corners.
The design of bearings for bridges with skew angles
less than 70 require consideration of out-of-plane
rotations.
AASHTO provides consideration of skew angle by way
of moment and shear correction factors. PennDOT agrees
with the application of the shear correction factors.
PennDOT has decided not to take advantage of the reduction
in load distribution factors for moment. However, these
factors do not adequately address problems due to out-ofplane rotations, uplift, or cross-frame forces. The provisions
in this section are meant to be applied to account for these
items. Note that uplift on concrete structures is not
considered as critical as that on steel structures.
During routine bridge inspections, the Department has
found many occurrences of buckled cross-frame members
and poor bearing performance on skewed structures. The
Department has found from refined analyses that crossframes in skewed structures are potentially subjected to
higher force levels than cross-frames in normal (90 )
structures. This does not mandate a 3-D analysis, but does
B.4 - 4
September 2007
SPECIFICATIONS
COMMENTARY
Steel structures with skew angles less than 70 require
a special cross-frame design and the cross-frame
members must be considered as main load carrying
members.
Table 1 describes how the term L (length) shall be
determined for use in the live load distribution factor
equations given in A4.6.2.2.2, A4.6.2.2.3, D4.6.2.2.2 and
D4.6.2.2.3.
mean a special analysis of the cross-frame must be provided
in order to account for the differential deflections which
occur across a cross-frame. This analysis should accurately
account for cross frame member geometry and stiffness.
Should a grillage analogy model be used, accurate
representation of cross frame stiffness should be established
via special analysis of representative frames. Should a more
sophisticated 3-D analysis be used, models can be
constructed following the technique recommended in
C4.6.1.2.1 for structures curved in plan.
A crude or approximate uplift check can be made using
the second term of the adjustment factor as an estimate of
negative live load reaction potential. Comparing the
negative reaction with the dead load reaction will provide an
estimate of the potential for uplift. The engineer must use
engineering judgement regarding the applicability of this
method.
An acute corner is defined as the corner of the structure
where the angle formed by intersection of edge of the deck
and the centerline of bearings is less than 90 .
Proper consideration of out-of-plane rotations during
the bearing design is also required. Normally out of plane
rotations will require multi-rotational bearings.
This method incorporated in this manual for
determining L seems to be appropriate for the level of
sophistication of the distribution factors. As additional
knowledge is gained on this subject, this method for
determining L may be modified.
B.4 - 5
SPECIFICATIONS
September 2007
COMMENTARY
Table 4.6.2.2.1-1 - L for Use in Live Load Distribution Factor Equations
CONDITION
FORCE EFFECT
L (mm) {ft.}
A
Positive Moment
The length of the span for which
moment is being calculated.
B
Negative Moment End spans of
continuous spans,
from end to point of
dead load
contraflexure
C
Negative Moment Near interior
supports of
continuous spans,
from point of dead
load contraflexure
to point of dead
load contraflexure
The average length of the two
adjacent spans.
D
Negative Moment Interior spans of
continuous spans,
from point of dead
load contraflexure
to point of dead
load contraflexure
E
Shear
shear is being calculated.
F
Exterior Reaction
The length of the exterior span.
G
Interior Reaction of
Continuous Span
The average length of the two
adjacent spans.
Figure 1 provides a graphical representation of the information
given in Table 1.
B.4 - 6
September 2007
Figure 4.6.2.2.1-1 - L for use in Live Load Distribution Factor Equations
B.4 - 7
SPECIFICATIONS
COMMENTARY
In the rare occasion when the continuous span
arrangement is such that an interior span does not have any
positive dead load moment (i.e., no dead load points of
contraflexure), the region of negative moment near the
interior supports would be increased to the centerline of the
span, and the L used in determining the live load
distribution factors would be the average of the two adjacent
spans.
4.6.2.2.2 Distribution Factor Method for Moment and Shear
4.6.2.2.2a Interior Beams with Wood Decks
The following shall supplement A4.6.2.2.2a.
The distribution factors given in Table A4.6.2.2.2a-1
for Glued Laminated Panels on Glued Laminated Stringers
are applicable for panels with a 150 mm {6 in.} minimum
nominal thickness.
4.6.2.2.2b Interior Beams with Concrete Decks
C4.6.2.2.2b
A4.6.2.2.2b.
For preliminary design, the terms K g/(Lt3) {Kg/(12Lt3)}
in Table 1 shall be taken as 1.0 for non-composite beams.
The following shall replace Table A4.6.2.2.2b-1.
The following shall supplement AC4.6.2.2.2b.
In Table A4.6.2.2.2b-1, in the Category "Concrete
Beams used in Multi-Beam Decks", the cross-section, Type
g (from Table A4.6.2.2.1-1), with option "if sufficiently
connected to act as a unit" has been removed from Table
D4.6.2.2.2b-1. This option has been removed because it has
been difficult to provide enough post-tensioning for the
composite box beams to act as a unit.
B.4 - 8
September 2007
Table 4.6.2.2.2b-1 – Distribution of Live Loads Per Lane for Moment in Interior Beams
Metric Units
Type of Beams
Applicable
Cross-Section
from Table
A4.6.2.2.1-1
Wood Deck on Wood
or Steel Beams
a, 1
Concrete Deck on
Wood Beams
Concrete Deck, Filled
Grid, or Partially
Filled Grid on Steel or
Concrete Beams:
Concrete T-Beams, Tand Double T-Sections
1
a, e, k
See Table A4.6.2.2.2a-1
One Design Lane Loaded:
S/3700
Two or More Design Lanes Loaded:
S/3000
S
1100
S
0.06 +
2930
d
1.0
1800
4900
300
152 000
If L > 73 000, use
L = 73 000
4x109 Kg 3.2 1012
Nb 4
0.2
S
L
Kg
0.1
Lt S3
S
3350
0.075 +
1.0
S
L
0.08
Kg
0.1
Lt S3
Use lesser of the above or Lever Rule
Nb = 3
4000
18 000 L 152 000
If L > 73 000, use
L = 73 000
Nc 3
S
1.75
1100
300
L
0.35
1
Nc
0.45
13
Nc
Concrete Deck on
Concrete Spread Box
Beams
Range of Applicability
ts
i, j
if sufficiently
connected t act
as a unit
Cast-in-Place Concrete
Multicell Box
Distribution Factors
b, c
0.3
S
430
1
L
0.25
One Design Lane Loaded
S
910
0.35
Sd
0.25
L2
S
1900
0.6
Sd
S
6400
If Nc > 8 use Nc = 8
For two or more design lanes
loaded
If L > 427 000/Nc, use
L = 427 000/Nc
1800
S 5500
L 152 000
If L > 43 000, use
L = 43 000
430 d 1700
Nb 3
0.125
L2
Use Lever Rule
S > 5500
B.4 - 9
September 2007
Table 4.6.2.2.2b-1 – Distribution of Live Loads Per Lane for Moment in Interior Beams (Continued)
Metric Units
Type of Beams
Concrete Beams used
in Multi-Beam Decks
Applicable
Cross-Section
From Table
A4.6.2.2.1-1
f
b
2.8L
k
0.5
I
J
g
Steel Grids on Steel
Beams
h
b
1500
152 000
If L 37 000, use
L = 37 000
5
Nb
20
0.25
where : k = 2.5(Nb)-0.2
if sufficiently
connected to act
as a unit
1500
152 000
Nb
20
If Nb 12, use Nb = 12
1.5
b
k
7600
0.6
b
L
0.2
I
J
0.06
5.66
Nb
15
L0.15
Regardless of Number of Loaded Lanes: S/D where:
C=K(W/L)
g, i, j
if connected only
enough to prevent
relative vertical
displacement at
the interface
D=300[11.5 – NL + 1.4NL (1 - 0.2C)2]
when C 5
D=300 {11.5 – NL} when C 5
K=
(1
)I
J
For preliminary design, the following values of K may
be used:
Beam Type
Non-voided rectangular beams
Rectangular beams with circular voids
Box section beams
Channel beams
T-beam
Double T-beam
Beams
Concrete deck on
Multiple Steel Box
Girders
a
b, c
K
0.7
0.8
1.0
2.2
2.0
2.0
S/2300 If tg< 100 mm
S/3050 If tg 100 mm
1800 mm
S/2400 If tg< 100 mm
S/3050 If tg 100 mm
1800 mm
Regardless of Number of Loaded Lanes:
0.05 + 0.85
NL
Nb
0.425
NL
B.4 - 10
0.5
NL
Nb
1 .5
September 2007
Type of Beams
Wood Deck on Wood
or Steel Beams
Concrete Deck on
Wood Beams
Grid, or Partially Filled
Grid on Steel or
Concrete Beams;
a, l
l
a, e, k
6.0′
S/12
S/10
S
16.0'
12"
500'
If L > 240', use L = 240'
10,000 in4 Kg 7,600,000 in4
Nb ≥ 4
1.0
S
0.06 +
9.6
S
L
0.2
Kg
0.1
12 Lt S3
d
S
11
0.075 +
1.0
S
L
0.08
Kg
0.1
12 Lt S3
Use lesser of the above or Lever Rule
Nb = 3
7.0’ S 13.0’
60’ L 500’
If L > 240’, use L = 240’
Nc 3
1.75
S
3.6
1
L
0.35
1
NC
0.45
13
NC
Concrete Deck on
Concrete Spread Box
Beams
ts
i, j
If sufficiently
connected to act
as a unit
Multicell Box
Applicable
Cross-Section
from Table
A4.6.2.2.1-1
b, c
0.3
S
5.8
1
L
0.25
S
3.0
0.35
Sd
0.25
12 L2
S
21
If Nc > 8 use Nc = 8
For two or more design lanes
loaded
If L > 1400/Nc, use
L = 1400/Nc
S
18.0'
500'
If L > 140', use L = 140'
d 66"
Nb 3
S
6.3
0.6
Sd
0.125
12 L2
S ≥ 18.0’
Use Lever Rule
B.4 - 11
September 2007
Type of Beams
Applicable
Cross-Section
from Table
A4.6.2.2.1-1
Concrete Beams
used in Multi-Beam
Decks
f
0.5
35" b 60"
20' L 500'
If L 120', use L = 120'
5 Nb 20
0.25
b
I
33.3L
J
where: k = 2.5(Nb)-0.2
k
g
if sufficiently
connected to act
as a unit
h
g, i, j
if connected
only enough to
prevent relative
vertical
displacement at
the interface
1.5
k
b
305
0.6
b
12 L
0.2
I
J
0.06
2.4
35" b 60"
20' L 500'
5 Nb 20
If Nb 12, use Nb = 12
Nb
15
L0.15
Regardless of Number of Loaded Lanes: S/D
where:
C=K(W/L)
D = 11.5 – NL + 1.4NL (1 – 0.2C)2
When C 5
D = 11.5 – NL when C > 5
K=
(1
)l
J
for preliminary design, the following values of
K may be used:
Beam Type
Non-voided rectangular beams
Rectangular beams with circular voids
Box section beams
Channel beams
T-beam
Double T-beam
Beams
a
S/7.5' If tg 4"
S/10.0' If tg 4"
S/8.0' If tg 4"
S/10.0' If tg 4"
Concrete deck on
Multiple Steel Box
Girders
b, c
Regardless of Number of Loaded Lanes:
N
0.425
0.05 0.85 L
Nb
NL
B.4 - 12
K
0.7
0.8
1.0
2.2
2.0
2.0
S
6.0’
S
6.0’
0.5
NL
Nb
1 .5
September 2007
SPECIFICATIONS
COMMENTARY
4.6.2.2.2d Exterior Beams
C4.6.2.2.2d
The following shall supplement AC4.6.2.2.2d.
The value of de is to be computed using the midpoint of
the exterior web.
4.6.2.2.2e Skewed Bridges
C4.6.2.2.2e
Delete A4.6.2.2.2e.
The following shall replace AC4.6.2.2.2e.
PennDOT has decided not to take advantage of the
reduction in load distribution factors for moment in
longitudinal beams on skewed supports.
4.6.2.2.3 Distribution Factor Method for Shear
4.6.2.2.3a Interior Beams
The following shall replace the second sentence of
Paragraph 1 in A4.6.2.2.3a.
For interior beams not listed in Table 1, lateral
distribution of axle load shall be determined by lever rule.
The following shall replace Table A4.6.2.2.3a-1.
B.4 - 13
Table 4.6.2.2.3a-1 – Distribution of Live Loads Per Lane for Shear in Interior Beams
Metric Units
Type of Superstructure
Applicable
Cross-Section
from Table
A4.6.2.2.1-1
One Design Lane
Loaded
Wood Deck on Wood or
Steel Beams
Concrete Deck on Wood
Beams
Grid, or Partially Filled
Grid on Steel or
Concrete Beams;
Multicell Box
Concrete Deck on
Concrete Spread Box
Beams
Concrete Beams Other
Than Box Beams Used
in Multi-Beam Decks
Steel Grid Deck on Steel
Beams
Concrete Deck on
Multiple Steel Box
Beams
l
a, e, k
I, j
if sufficiently
connected to act
as a unit
d
b, c
Lever Rule
Lever Rule
S
7600
0.36
Lever Rule
S
2900
S
3050
f, g
0.70
H
b
L
S
10 700
2
Lever Rule
0.6
0.6
N/A
S
0. 2
3600
0.1
d
L
0.1
d
L
Lever Rule
Concrete Box Beams
Used in Multi-Beam
Decks
Two or More Design
Lanes Loaded
0.15
S
2200
S
2250
0.9
0.8
Nb = 3
d
L
d
L
1800
0.1
S > 5500
b
0.05
b
4000
0.4
b
L
0.1
l
J
0.05
1500
L 152 000
If L > 37 000, use
L = 37 000
5 Nb 20
6.8x109 J 3.75x1011
2.1x109 I 3.75x1011
Lever Rule
Lever Rule
N/A
Lever Rule
Lever Rule
N/A
b, c
As specified in Table D4.6.2.2.2b-1
B.4 - 14
4000
152 000
2800
S 5500
L 152 000
If L > 43 000, use
L = 43 000
d 1700
Nb 3
0.1
i, j
if connected only
enough to
prevent relative
vertical
displacement at
the interface
a
S
L
d
Nc ≥ 3
Lever Rule
I
J
1100 S 4900
6000 L 152 000
If L > 73 000, use
L = 73 000
ts 300
Nb 4
Table 4.6.2.2.3a-1 – Distribution of Live Loads Per Lane for Shear in Interior Beams (Continued)
Type of
Superstructure
Applicable
Cross-Section
from Table
A4.6.2.2.1-1
One Design Lane
Loaded
Wood Deck on
Wood or Steel
Beams
Concrete Deck on
Wood Beams
Concrete Deck,
Filled Grid, or
Partially Filled Grid
on Steel or Concrete
Beams; Concrete TBeams, T- and
Double T-Sections
Two or More Design
Lanes Loaded
l
a, e, k
Lever Rule
Lever Rule
S
25
0.36
0.2
S
12
N/A
S
35
2
3.5'
S 16.0'
L 500'
If L > 240', use L = 240'
ts 12"
Nb 4
0.1
6.0' S 13.0'
20 L 500'
110"
Nc 3
0.1
S 18.0'
L 500'
If L > 140', use L = 140'
d 66"
Nb 3
i, j
If sufficiently
connected to act
as a unit
Cast-in-Place
Concrete Multicell
Box
d
Concrete Deck on
Concrete Spread
Box Beams
b, c
S
9.5
S
10
0.6
0.6
d
12 L
0.1
0.1
d
12 L
S
7.4
Lever Rule
Concrete Box
Beams Used in
Multi-Beam Decks
f, g
Concrete Beams
Other Than Box
Beams Used in
Multi-Beam Decks
h
b
130 L
0.15
S
7.3
0.9
0.8
d
12 L
d
12L
S > 18.0’
Lever Rule
l
J
0.05
b
156
0.4
b
12 L
0.1
l
J
0.05
b 60"
L 500'
If L > 120', use L = 120'
5 Nb 20
16,500 in4 J 900,000 in4
5,000 in4 I 900,000 in4
Lever Rule
Lever Rule
N/A
Lever Rule
Lever Rule
N/A
i, j
if connected only
enough to prevent
relative vertical
displacement at
the interface
Steel Grid Deck on
Steel Beams
a
Concrete Deck on
Multiple Steel Box
Beams
b, c
As specified in Table D4.6.2.2.2b-1
B.4 - 15
September 2007
SPECIFICATIONS
COMMENTARY
4.6.2.2.3c Skewed Bridges
C4.6.2.2.3c.
A4.6.2.2.3c.
In determining end shear for beams other than
prestressed concrete adjacent box beams, the shear skew
adjustment factor shall be applied to the shear distribution
factor of exterior beams at the obtuse corners for a distance
of one-half the span length (see Figure 1).
In determining end shear for prestressed concrete
adjacent box beams, the shear skew adjustment factor shall
be applied to the shear distribution factors for all the beams
which are on a skew (see Figure 1).
The following shall supplement AC4.6.2.2.3c.
When structures have multiple skew angles, the
smallest applicable skew angle associated with that girder
should be used in calculation of shear correction factors.
Figure 4.6.2.2.3c-1 - Application of Shear Correction Factor for End Shear
In determining end reactions of continuous beams (such
as reactions at abutments) for beams other than prestressed
concrete adjacent box beams, the shear skew adjustment
factor shall be applied to the shear distribution factor of
exterior beams at the obtuse corners (see Figure 2).
In determining end reactions of continuous beams (such
as reactions at abutments) for prestressed concrete adjacent
box beams, the shear skew adjustment factor shall be
B.4 - 16
September 2007
SPECIFICATIONS
COMMENTARY
applied to the shear distribution factors for all the beams
which are on a skew (see Figure 2).
Figure 4.6.2.2.3c-2 - Application of Shear Skew Correction Factor for End Reactions
B.4 - 17
September 2007
SPECIFICATIONS
COMMENTARY
The following shall replace Table A4.6.2.2.3c-1.
Table 4.6.2.2.3c-1 - Correction Factors for Load Distribution Factors for Support Shear of the Obtuse Corner
Metric Units
Concrete Deck, Filled Grid,
or Partially Filled Grid on
Steel or Concrete Beams;
Concrete T-Beams, T- and
Double T-Section
Multicell Box
Correction Factor
Applicable CrossSection from
Table 4.6.2.2.1-1
a, e, k and also i, j
if sufficiently
10
. 0.20
connected to act as a
unit
d
Concrete Deck on Spread
Concrete Box Beams
b, c
Concrete Box Beams Used in
Multibeam Decks
f, g
L ts3
Kg
0.25
10
.
Ld
tan 90
6S
L
tan
90d
B.4 - 18
tan 90
L
tan 90
70d
10
.
10
.
0 .3
30º ≤ ≤ 90º
≤ ≤ 4900
≤ ≤ 152 000
If L > 73 000, use
L = 73 000
Nb ≥ 4
≤ ≤ 90º
≤ 4000
≤ ≤ 152 000
If L > 73 000, use
L = 73 000
≤ ≤ 2700
Nb ≥ 3
30º ≤ ≤ 90º
≤ ≤ 3500
If S > 3500, use S = 3500
≤ ≤ 152 000
If L > 43 000, use
L = 43 000
≤ ≤ 1700
Nb ≥ 3
≤ ≤ 90º
≤ ≤ 152 000
If L > 37 000, use
L = 37 000
≤ ≤ 1700
≤ ≤ 1500
≤ Nb ≤ 20
September 2007
SPECIFICATIONS
COMMENTARY
Table 4.6.2.2.3c-1 - Correction Factors for Load Distribution Factors for Support Shear of the Obtuse Corner (Continued)
Concrete Deck, Filled Grid,
or Partially Filled Grid on
Steel or Concrete Beams;
Concrete T-Beams, T- and
Double T-Section
Multicell Box
Concrete Deck on Spread
Concrete Box Beams
Correction Factor
Applicable
Cross-Section from
Table 4.6.2.2.1-1
a, e, k and also i, j
12 L t s 3
if sufficiently
10
. 0.20
Kg
connected to act as a
unit
d
10
.
0.25
b, c
Concrete Box Beams Used in
Multibeam Decks
f, g
12 L
90d
10
.
4.6.2.3 EQUIVALENT STRIP WIDTHS FOR SLABTYPE BRIDGES
A4.6.2.3.
This article shall be applied to the types of crosssections which are shown schematically in Table 1. For the
purpose of this article, cast-in-place voided slab bridges may
be considered as slab bridges.
4.6.2.5 EFFECTIVE LENGTH FACTOR, K
For Extreme Event I, Seismic Loading, the effective
length factor, K, in the plane of bending may be assumed to
be unity in the calculation of λ.
B.4 - 19
tan 90
12 L
tan 90
70 d
Ld
12.0
tan 90
6S
10
.
0.3
tan 90
30º θ 90º
3.5' S 16.0'
20' L 500'
If L > 240', use L = 240'
Nb 4
30º θ 90º
6.0' < S 13.0'
20' L 500'
If L > 240', use L = 240'
35" d 110"
Nb 3
30º θ 90º
6.0' S 11.5'
If S > 11.5', use S = 11.5'
20' L 500'
If L > 140', use L = 140'
17" d 66"
Nb 3
30º θ 90º
20' L 500'
If L > 120', use L = 120'
17" d 66"
35" b 60"
5 Nb 20
September 2007
SPECIFICATIONS
COMMENTARY
4.6.2.6 EFFECTIVE FLANGE WIDTH
4.6.2.6.1 General
C4.6.2.6.1
The following shall replace the last two sentences of the
first paragraph.
For the calculation of live load deflections, where
required, the provisions of D2.5.2.6.2 shall apply.
For typical continuous span bridges, the effective span
length may be taken as the distance between points of dead
load contraflexure, and can be approximated as shown in
Table 1 below:
Table C4.6.2.6.1-1 - Approximate Effective Span Lengths for
Computing Effective Slab Width of Continuous Span Bridges
Positive Flexure
Regions
Exterior
Span
Interior
Span
0.7 LE
0.5 LI
Negative Flexure Regions
Support
Not
Adjacent to
Exterior
Span
For
Two-Span
Girder
0.25 LI1 +
0.25 LI2
0.3 LE1 +
0.3 LE2
Support
Adjacent to
Exterior
Span
0.30 LE +
0.25 LI
Where LI and LE represent interior and exterior span lengths,
respectively. This approximation is only applicable for the
purpose of computing effective span lengths to be used in
the determination of the effective slab width.
The
approximation is based on assumed points of dead load
contraflexure for balanced span lengths. In cases where this
simplification is not valid, i.e., the spans are not balanced, a
more rigorous estimation of the effective span length should
be made by determining the actual points of dead load
contraflexure. Because of the effective slab width affects
the section properties to be used in the girder dead load
analysis, an iterative procedure may be required to
determine the appropriate values. In such cases, a 5%
tolerance is allowed on the effective slab width, bs.
Where the cross-section of the web or top flange
changes over a flexural region, the minimum top flange
width or web thickness within that region may be used in
the calculation of the effective slab width.
Following the determination of the actual points of dead
load contraflexure by the continuous beam analysis, the
designer should compute the effective slab width in
accordance with A4.6.2.6.1 based on the true effective span
length. If the effective span length controls the calculation
of bs, the calculated value should be compared to the
previously computed value to determine whether it is within
the allowed tolerance. The effective span length will
typically control the calculation of bs only for short-spans
having large girder spacings.
For steel girders, an acceptable approximation of the
permanent load inflection points is the non-composite dead
load contraflexure points based on the non-composite
B.4 - 20
September 2007
SPECIFICATIONS
COMMENTARY
section properties.
For prestressed concrete girders, an acceptable
approximation of the permanent load inflection points is the
composite dead load contraflexure points based on the
composite (3n) section properties.
4.6.2.10P GIRDER - FLOORBEAM - STRINGER
BRIDGES
4.6.2.10.1P Girder Live Load Distribution Factors
Girder live load distribution factors shall be calculated
on the assumption that deck acts as a simple span between
the girders or deck acts as beam with overhang for exterior
girders (i.e., this assumes the stringers and floorbeams are
not present).
4.6.2.10.2P Stringer Live Load Distribution Factors
Stringer live load distribution factors shall be based on
D4.6.2.2 and A4.6.2.2.
4.6.2.10.3P Floorbeam Live Load Distribution Factors
4.6.2.10.3aP Floorbeams with the Top Flange not Directly
Supporting the Deck
For floorbeams with the top flange not directly
supporting the deck, the longitudinal reaction of design live
load is determined and then these loads are moved
transversely along the floorbeam to produce the maximum
force effect assuming the stringers are not present.
4.6.2.10.3bP Floorbeams with the Top Flange Directly
Supporting the Deck
For floorbeams with the top flange directly supporting
the deck, the floorbeam distribution shall be calculated as
given in A4.6.2.2.
4.6.2.11P DISTRIBUTION OF LOAD FROM THE
SUPERSTRUCTURE TO THE SUBSTRUCTURE
C4.6.2.11P
In order to determine girder reactions (which are used
as loads for the substructure design), the deck is assumed to
act as a simple beam between interior girders and as a
cantilever beam for the exterior girder and the first interior
girder.
In the calculation of live load girder reactions, the
design vehicle shall be assumed to spread uniformly over
3000 mm {10 ft.} (i.e., not two separate wheel loads).
For abutments designed on a per meter basis, an
acceptable alternate method of distribution would be to
divide the sum total of all the loads applied to the abutment
(for each limit state) by the abutment front face width.
When using the above approach, the live load contribution
may be obtained by determining the live load effect for one
lane of loading and multiplying that effect by the number of
design lanes.
B.4 - 21
September 2007
SPECIFICATIONS
COMMENTARY
4.6.2.12P EQUIVALENT STRIP WIDTHS FOR BOX
CULVERTS
C4.6.2.12.1P
Design for depths of fill of 600 mm {2.0 ft}or greater
are covered in D3.6.1.2.6
4.6.2.12.1P General
This article shall be applied to box culverts with depths
of fill less than 600 mm {2.0 ft}.
4.6.2.12.2P Case 1: Traffic Travels Parallel to Span
C4.6.2.12.2P
When traffic travels primarily parallel to the span,
culverts shall be analyzed for a single loaded lane with the
single lane multiple presence factor.
Culverts are designed under the provisions of Section
D12.
Box culverts are normally analyzed as twodimensional frames. Equivalent strip widths are used to
simplify the analysis of the three-dimensional response to
live loads. Equations 1 and 2 are based on research
(McGrath et al., 2004) that investigated the forces in box
culverts with spans up to 7300 mm {24 ft}.
The axle load shall be distributed to the top slab for
determining moment, thrust, and shear as follows:
Perpendicular to the span:
Metric Units:
E = 2440 + 0.12 S
(4.6.2.12.2P-1)
E = 96 + 1.44 S
Parallel to the span:
Espan = LT + LLDF (H)
(4.6.2.12.2P-2)
The distribution widths are based on distribution of
shear forces. Distributions widths for positive and negative
moments are wider; however, using the narrower width in
combination with a single lane multiple presence factor
provides designs adequate for multiple loaded lanes for all
force effects.
Although past practice has been to ignore the
distribution of live load with depth of fill, consideration of
this effect, as presented in Equation 2, produces a more
accurate model of the changes in design forces with
increasing depth of fill. The increased load length parallel
to the span, as allowed by Equation 2, may be
conservatively neglected in design. For the BXLRFD
computer program, Equation 2 has been ignored.
where:
E
=
equivalent distribution width perpendicular to
span (mm){in.}
S
=
clear span (mm) {ft.}
Espan =
equivalent distribution length parallel to span
(mm) {in.}
LT
=
length of tire contact area parallel to span, as
specified in D3.6.1.2.5 (mm) {in.}
LLDF =
factor for distribution of live load with depth
of fill, 1.00, as specified in D3.6.1.2.6
H
depth of fill from top of culvert to top of
pavement (mm) {in.}
=
4.6.2.12.3P Case 2: Traffic Travels Perpendicular to Span
C4.6.2.12.3P
When traffic travels perpendicular to the span,
distribute live load to the top slab using the equations in
Culverts with traffic traveling perpendicular to the span
can have two or more trucks on the same design strip at the
B.4 - 22
September 2007
SPECIFICATIONS
A4.6.2.1 for concrete decks with
perpendicular to the direction of traffic.
COMMENTARY
primary
strips
same time. This must be considered, with appropriate
multiple presence factor, in analysis of the culvert structural
response.
4.6.2.12.4P Precast Box Culverts
C4.6.2.12.4P
For precast box culverts, the distribution width
computed with Equation 4.6.2.12.2P-1 shall not exceed the
length between two adjacent joints without a means of shear
transfer across the joint. Additionally, if no means of shear
transfer is provided, the section ends shall be designed as an
edge beam in accordance with the provisions of A4.6.2.1.4b.
Shear transfer may be provided by pavement, soil
backfill, or a physical connection between adjacent sections.
Precast box culverts manufactured in accordance with
AASHTO Materials Specification M273 are often installed
with joints that do not provide a means of direct shear
transfer across the joints of adjacent sections under service
load conditions. This practice is based on research (James,
1984, Frederick, et al., 1988) that showed small deflections
and strains under wheel loads with no earth cover, due
primarily to the fact that the sections were designed as
cracked sections but do not crack under service loading.
While there are no known service issues with installation of
standard box sections without means of shear transfer across
joints, analysis (McGrath et al., 2004) shows that stresses
are substantially higher when a box culvert is subjected to a
live load at a free edge than when loaded away from a free
edge.
Most shallow cover box culvert applications have some
fill or a pavement that likely provide sufficient shear
transfer to distribute live load to adjacent box sections
without shear keys to avoid higher stresses due to edge
loading. States and design agencies that utilize grouted shear
keys, pavement or systems whose function is to transfer
shear across joints may use past performance of these
connections and/or materials as a basis for providing
adequate shear transfer. Otherwise, for applications with
zero depth of cover, and no pavement, soil, or other means
of shear transfer such as shear keys, designers should design
the culvert section for the specified reduced distribution
widths. The use of post-tensioning in accordance with the
BC Standard Drawings in conjunction with a cast-in-place
slab or bituminous pavement is considered sufficient to
provide adequate shear transfer between adjacent culvert
sections.
B.4 - 23
September 2007
SPECIFICATIONS
COMMENTARY
4.6.3 Refined Methods of Analysis
4.6.3.1 GENERAL
C4.6.3.1
A4.6.3.1.
Barriers shall not be considered in the calculation of the
structural stiffness nor structural resistance of a structure.
When a refined method of analysis is performed for
beam-slab bridges other than those bridges defined in
D4.6.1.2.1 and D4.6.2.2.1 (which must use a refined method
analysis), the beams must have capacity not less than if it
were designed using the approximate method of analysis
given in A4.6.2.2.1 and D4.6.2.2.1.
Delete the second paragraph of AC4.6.3.1.
4.6.3.2 DECKS
4.6.3.2.3 Orthotropic Plate Model
In orthotropic plate modeling, the flexural rigidity of
the elements may be uniformly distributed along the crosssection of the deck. Where the torsional stiffness of the
deck is not contributed solely by a solid plate of uniform
thickness, the torsional rigidity should be established by
physical testing, three-dimensional analysis, or generally
accepted and verified approximations, and shall be approved
by the Chief Bridge Engineer.
4.6.3.3 BEAM-SLAB BRIDGES
C4.6.3.3
When a refined method of analysis is performed, the
live load force effects carried by each girder may be
computed by the techniques listed below in order of
decreasing sophistication:
The following shall supplement the bulleted list of
AC4.6.3.3.
Three-dimensional finite element method, and
Two-dimensional grillage analogy
With a two-dimensional grillage analogy or a threedimensional finite element method, the rating of a bridge is
not as straightforward as when the A4.6.2.2 or D4.6.2.2 is
used. A refined method of analysis more closely represents
the fact that the distribution of live loads on a bridge is not
described by a constant distribution factor. When a refined
method of analysis is used in the design, a table of live load
distribution factors (based on design truck of the PHL-93)
for girder maximum positive and negative moments and
shear in each span shall be provided on the contract plans to
aid in future ratings of the bridge. The live load distribution
factor shall be in the form of a ratio of the force effect from
the refined method of analysis caused by the design truck of
the PHL-93 in that lane, divided by the force effect obtained
B.4 - 24
If the program being used allows only for nodal loads,
concentrated loads shall be distributed to adjacent nodes
by simple statics. If the spacing of girders significantly
exceeds 2400 mm {8 ft.}, it is preferable to place
intermediate nodes on the transverse members to model
load distribution more accurately.
The framing of members at bearings is very important.
Nodes at bearings should not be artificially restrained
through the enforcement of fixed support conditions for
other than vertical transitional support at all bearings,
and longitudinal and transverse transitional support
where the detailing dictates. This provision is critical to
proper modeling of bridges with significant skew.
The warping stiffness of girders is difficult to model,
unless a warping (not St. Venant torsional) degree of
freedom is explicitly provided by the computer
program. As an approximation, the dead load moment
diagram and live load moment envelope may be
converted to a uniform flange bending (i.e., bimoment)
September 2007
SPECIFICATIONS
COMMENTARY
from application of one design truck of the PHL-93 acting
on a single, isolated girder. The commentary provides an
example table of live load distribution factors.
Two other areas which must be considered for future
ratings of bridges designed with refined method analysis are
cross-frame forces and uplift at reactions. Therefore, at
Type, Size and Location submission, the designer must
include (for approval by the Chief Bridge Engineer) a
proposed simplified method for rating a special vehicle for
the controlling cross-frame member and uplift reaction
condition.
When a grillage analogy or finite element model is
implemented, the preferred procedure is the generation and
subsequent loading of an influence surface to produce
maximum effect. The influence surface shall be loaded to
maximize positive and negative design values (moments,
shear, diaphragm forces, etc.) for all critical points along the
bridge. This process is analogous to the classical use of
influence lines. The provisions of A3.6.1.1.2 shall apply to
the loading method described above. It is also acceptable,
though less rigorous and economical, to apply distribution
factors from D4.6.2.2.2 to each girder in a grid system and
allow the grid or finite element analysis to more accurately
determine the load distribution to interior and exterior
beams, and to account for the influences of skew and
curvature, if present. The latter procedure is less preferred
than the former procedure since uneconomical designs may
result.
It is expressly not acceptable to load a grid or finite
element model with the number of design lanes of live load
given in A3.6.1.1.1, without simultaneously positioning the
loads longitudinally and transversely to maximize and
minimize each design moment, shear, diaphragm load
(curved and/or skewed bridges), and reaction. Except in the
most obvious cases (e.g., bridges that are essentially right
and straight), the transverse and longitudinal position of live
load which produces the critical design condition should be
determined by the use of influence surfaces.
loading by the approximation below:
M
Rd
W
where:
ΣM =
total primary moment (N mm) per mm
{k in/ft}
R
=
radius (mm) {ft.}
d
=
girder depth (mm) {in.}
This load may be applied to the flange and assumed to
be reacted by the diaphragms and slab, as appropriate,
analogous to a continuous beam on simple supports. If
a warping degree of freedom or some approximate
method of calculating warping stress is not included in
the program, it will be necessary to combine the loads
calculated according to the above method with the loads
produced from diaphragms.
If grid or finite element methods are used to determine
dead load forces, then the structure, including
diaphragms, must be cambered in accordance with the
computer analysis. If some cambering other than that
corresponding to the opposite of the natural dead load
deflection is anticipated (e.g., cambering for more
efficient dead load distribution), it must be reflected in
the computer modeling.
Special care shall be exercised in modeling, analysis
and interpretation of the results of two- and threedimensional procedures. AC4.6.3.3 and this commentary
provide guidelines. The designer is responsible for the
correct application of advanced analysis methods and is
advised that various commercial and generic computer
programs can report significantly different results for
various combinations of skew and/or curvature.
Figure C1 provides an example of a table for live load
distribution factors which shall be included on the contract
drawings. Since the figure provides only example tables,
the table used on the contract drawings shall be developed
for the specific structure in question.
When a refined method of analysis is used to design a
bridge, the table of live load distribution factors shall be the
basis for future ratings of the bridge. In order to provide a
realistic rating in the absence of such a table, the rating of
the bridge would require modeling the bridge by means of
an analysis method similar to the one used in the original
design.
The live load distribution factors given in
D4.6.2.2.2 should not be used for the rating, since they
would generally provide an unduly conservative rating or, in
a few cases, an unconservative rating.
B.4 - 25
September 2007
SPECIFICATIONS
COMMENTARY
Girder 1-Moment Distribution Factors*
Positive Moment
Location
Negative Moment
1 Lane
2 Lanes
N Lanes
Span 1
Interior
Support 1
–
–
–
–
–
–
N Lanes
–
–
–
–
–
–
–
–
–
–
Span N
Interior
Support N
2 Lanes
–
Span 2
Interior
Support 2
1 Lane
–
Girder 1 – Shear Distribution Factors*
Location
1 Lane
2 Lanes
N Lanes
*These tables shall be repeated for each girder
Figure C4.6.3.3-1 - Example of Live Load Distribution Factor Tables
A live load distribution factor developed for the design
truck may be used for permit and rating vehicles. When the
bridge or the vehicle or both are unusual, an analysis should
be made to justify the use of an design truck distribution
factor for other types of vehicles.
One possible way to develop a simplified method for
rating for cross-frame members would be to develop a table
of ratings for PHL-93, P-82, ML-80, HS20 and H20 for 1 up
to N lanes loaded at the time of original design. When a
future rating of a special vehicle is required, the engineer
could develop an approximate rating by interpolating among
these vehicles used to develop the table.
A similar approach could be used for the uplift reaction
condition. A table of uplift reactions for PHL-93, P-82,
ML-80, HS20 and H20 for 1 up to N lanes loaded could be
developed at the time of original design. When a future
rating of a special vehicle is required, the engineer could
develop an approximate uplift reaction by interpolating
among these vehicles used to develop the table. Next, this
uplift reaction could be in computing a rating.
Some programs use an influence surface loading
technique that evaluates numbers of lanes of loading with
B.4 - 26
SPECIFICATIONS
September 2007
COMMENTARY
appropriate multi-presence reduction factors.
Other
procedures apply the distribution factors given in D4.6.2.2.2
to each girder in a grid of girders and diaphragms and allow
a grid analysis to evaluate the effects of skew, curvature,
and the tendency for load to gravitate toward the exterior
stringers, but they do not necessarily take advantage of the
fact that the true distribution factor is often less than those
given in D4.6.2.2.2 and typically is not constant for all
effects (moment, shear, deflection, etc.) or locations along
the bridge.
In 1986, the Department conducted a parametric study
of steel girder bridges to compare the results obtained from
these programs for a variety of combinations of skew and
curvature. This parametric study could not cover all cases
that might be encountered by designers, but did reveal some
trends and comparative data which form the basis of a report
titled "Review of Computer Programs for the Analysis of
Girder Bridges", January 1989. Since 1986, many of the
programs used in this study have been changed and/or
updated. Therefore, some of the report's comparisons and
observations may have lost their relevancy. For a copy of
this report, contact the Chief Bridge Engineer's office.
The modeling of diaphragms and boundary conditions
at supports and bearings is vital to obtaining the proper
results from some of these sophisticated programs. The
burden of correctly handling these factors rests squarely
upon the shoulders of the designer. Consider the following
example, which shows how a very small modeling error
produces very erroneous results.
The framing plan shown in Figure C2 represents a real
bridge that was designed using a grid-type approach. The
designer had a good model for this structure, except that the
rotational degree of freedom corresponding to the global xaxis at all of the bearings was fixed. This did not allow the
diaphragms at the piers and abutments to respond correctly
to the imposed loadings and deformations, and also had the
effect, by virtue of vector resolution between global and
local systems, of producing artificially stiff ends on the
girders.
This incorrect boundary assumption altered both the
dead load and live load forces. The effect of this on the
dead load reactions obtained at the abutments and piers was
dramatic. A modest uplift was reported at the near abutment
dead load reaction for Girder 1, and a very substantial uplift
was reported at the far abutment dead load reaction for
Girder 5. This is shown in the table in Figure C2, as is a
moment diagram for noncomposite dead loads which
reflects the incorrect dead load reactions. Also shown are
the correct dead load reactions which were determined when
the structure was modeled using the generic STRESS
computer program, with proper boundary conditions at the
supports. In this case, a positive dead load reaction was
found at all bearings, and a significantly different moment
diagram for noncomposite dead load also resulted. Finally,
the reaction table shows a set of incorrect dead load
reactions obtained with STRESS when same error in
B.4 - 27
SPECIFICATIONS
September 2007
COMMENTARY
boundary conditions was made to confirm the significance
of proper modeling. The correct and incorrect dead load
moment diagrams also are shown in Figure C2.
The differences in the degrees of freedom at the lines of
support on this structure were also investigated, utilizing a
relatively complete three-dimensional finite element
analysis and the SAPIV computer program. The model is
illustrated in Figure C3, which shows how the deck slab,
girders and cross-frames were modeled in their proper
relative positions in cross-section by means of rigid linking
members. Also shown in this figure is a comparison of the
dead load reactions obtained from STRESS and from
SAPIV by applying all the noncomposite dead loads in a
single loading. The agreement between these reactions is
excellent.
In order to verify that the order of pouring the deck slab
units would not contribute to an uplift situation, the pouring
sequence was replicated in a three-dimensional SAPIV
analysis. The results of the analysis of the three stages of
the pouring sequence are shown in Figure C3, as well as the
total accumulated load at the end of the pour. A comparison
of the sequential loading with the application of a single
loading of noncomposite dead load also showed relatively
good agreement.
There are cases in which dead load uplift at the
reactions due to skew and/or curvature is possible. The
simple span bridge featured in the November 1, 1984, issue
of Engineering News-Record was correctly analyzed and
indicated high corner dead load uplift.
The important point shown in the example in
Figures C2 and C3 is that seemingly small errors in
modeling the structure can result in very substantial changes
in the reactions, shears and moments. The designer must be
aware of this potential when using two- or threedimensional analysis techniques.
Sometimes modeling problems occur because user's
manuals are not clear or because a "bug" exists, of which the
author/vendor is unaware. Such a case is illustrated for a
simply supported, partially curved and skewed bridge in
Figure C4. Initially, this bridge was modeled with extra
joints at locations other than diaphragms in an effort to
improve live load determination. As a result, the number of
points along each girder was not equal, but there was no
indication of a potential problem in the descriptive
literature. The resulting live load moment envelopes for the
middle and two exterior girders, which are shown in
Figure C4a, are obviously unusual in shape and possibly
also in their order of maximum moment, i.e., No. 4, No. 5
and No. 3.
After these results were brought to the attention of the
authors/vendors, it was decided that the live load processor
was not responding properly to the unequal number of nodes
per girder, that nodes should be essentially "radial," and that
it was uncertain whether the nodes to which diaphragms
were not connected were legitimate. The revised model,
shown in Figure C4b, produced clearly better results, as is
B.4 - 28
SPECIFICATIONS
September 2007
COMMENTARY
shown in the indicated live load moment envelopes. This
example illustrates the need to review the output carefully
and to communicate with the authors/vendors about the
correct use of their programs.
In summary, two- and three-dimensional methods can
provide better understanding of load distribution in girder
bridges, but must be used with great care.
B.4 - 29
September 2007
Figure C4.6.3.3-2 - Framing Plan, Comparative Dead Load Reactions and Moment Diagram Showing Effect of Proper and
Improper Rotational Boundary Condition as Reflected in a Grid Analysis
B.4 - 30
September 2007
Figure C4.6.3.3-2 - Framing Plan, Comparative Dead Load Reactions and Moment Diagram Showing Effect of Proper and
Improper Rotational Boundary Condition as Reflected in a Grid Analysis (Continued)
B.4 - 31
September 2007
Figure C4.6.3.3-3 - Finite Element Idealization and Reactions Obtained for Structure Shown in Figure C4.6.3.3-1
B.4 - 32
September 2007
Figure C4.6.3.3-3 - Finite Element Idealization and Reactions Obtained for Structure Shown in Figure C4.6.3.3-1
(Continued)
B.4 - 33
September 2007
Figure C4.6.3.3-4 - Comparative Live Load Moment Envelopes for the Middle and Two Outside Girders of Curved Skewed
Showing the Results of an Apparent "Bug" in Algorithm for Applying Live Load
B.4 - 34
September 2007
Figure C4.6.3.3-4 - Comparative Live Load Moment Envelopes for the Middle and Two Outside Girders of Curved Skewed
Showing the Results of an Apparent "Bug" in Algorithm for Applying Live Load (Continued)
B.4 - 35
September 2007
SPECIFICATIONS
COMMENTARY
4.6.4 Redistribution of Negative Moments in Continuous
Beam Bridges
4.6.4.1 GENERAL
The redistribution of force effects in multi-span, multibeam or girder superstructures is not permitted in the design
of Pennsylvania bridges, except as described in D4.6.4.3.
4.6.4.2 REFINED METHOD
Delete A4.6.4.2
4.6.4.3 APPROXIMATE PROCEDURE
The following shall replace A4.6.4.3
A simplified redistribution procedure for compact steel
beams given in A6.10.2.2 and D6.10.2.2 may be used.
4.7 DYNAMIC ANALYSIS
4.7.1 Basic Requirements of Structural Dynamics
C4.7.1.4 DAMPING
Damping values obtained from field measurement or
tests shall be approved by the Chief Bridge Engineer.
4.7.2 Elastic Dynamic Responses
4.7.2.2 WIND-INDUCED VIBRATION
4.7.2.2.1 Wind Velocities
The Chief Bridge Engineer will decide if wind tunnel
tests are warranted for a structure.
4.7.4 Analysis for Earthquake Loads
4.7.4.3 MULTI-SPAN BRIDGES
C4.7.4.3.1
4.7.4.3.1 Selection of Method
The selection of the method of analysis depends on
seismic zone, regularity, and importance of the bridge.
Regularity is a function of the number of spans and the
distribution of weight and stiffness. Regular bridges have
less than seven spans, no abrupt or unusual changes in
weight, stiffness, or geometry; and no large changes in these
parameters from span-to-span or support-to-support,
abutments excluded. A more rigorous analysis procedure
may be used in lieu of the recommended minimum.
As a clarification of seismic analysis procedures, the
Department will accept a multi-mode analysis in all cases.
A single-mode or a multi-mode analysis is acceptable
for bridges whose skew angle is greater than 70º with a total
span length less than 152 000 mm {500 ft.}. For major and
unusual bridges, and bridges whose skew angle is less than
70º, use the multi-mode spectral analysis. Details of the
single-mode spectral method of analysis procedure are given
in A4.7.4.3.2.
Use the SEISAB computer program for seismic
analysis. In addition, structural analysis programs (such as
B.4 - 36
September 2007
SPECIFICATIONS
COMMENTARY
STAAD-III) may be used in lieu of SEISAB for unusual
structures, for structures having substructure units with
multi-column pier bents, and for structures with multiple
simple spans.
Detailed seismic analysis is not required for single span
bridges or for bridges in Seismic Zone 1. A seismic analysis
may be performed for multi-span bridges in Seismic Zone 1
if designers feel such an analysis will more accurately
reflect the connection forces and produce a more
economical design.
However, a multi-mode analysis is required when the bridge
skew angle is less than 70º or the bridge is a major (total of
span length greater than 152 000 mm {500 ft.}) or unusual
structure. For other cases, a single-mode analysis will
suffice.
SEISAB should be used for the seismic analysis. If
SEISAB is not suitable for a structure, other structural
analysis programs capable of seismic modeling (such as
STAAD-III) may be used in lieu of SEISAB. The Designer
must stipulate the reasons SEISAB is not suitable and obtain
the Department's approval.
It has been found that, in most cases, seismic forces do
not control the design.
Therefore, when modeling
foundations, it is acceptable for designers to perform a
number of analyses (computer runs), based on assumptions
which sufficiently bound the potential foundation stiffness
and then use the worst resulting forces and displacements to
design the structure. Use of this modeling procedure is
generally acceptable when seismic forces do not control the
design.
When seismic forces do control the design, an analysis
which more closely models the superstructure and
substructure stiffness is required.
Table 4.7.4.3.1-1 – Analysis Procedure
Seismic Zone
Bridges with Two or More Spans
1
None Required
2
A multi-mode analysis for all
major and unusual bridges, and
bridges whose skew angle is less
than 70º. A single mode (or a
multi-mode spectral) for bridges
whose skew angle is greater than
70º, with total span length less
than 152 000 mm {500 ft.}.
4.7.4.3.5P Determination of Elastic Forces and
Displacements
C4.7.4.3.5P
For multiple span bridges in Seismic Zone 2, the elastic
forces and displacements shall be determined independently
along two perpendicular axes by use of a single-mode or
multi-mode spectral method of analysis. The resulting
forces shall then be combined as specified in A3.10.8.
Typically the perpendicular axes are the longitudinal and
transverse axes of the bridge, but the choice is open to the
designer. The longitudinal axis of a curved bridge may be a
chord connecting the two abutments. For single span
bridges and bridges in Seismic Zone 1, the elastic forces and
displacements shall also be combined as specified in
A3.10.8.
When designing a component, it is important to
recognize its geometric restraints and/or releases, the
direction of the forces and displacements it must
accommodate, as well as the skew effects.
4.7.4.4 MINIMUM DISPLACEMENT REQUIREMENTS
C4.7.4.4
The following shall replace the definition of N in
Equation A4.7.4.4-1.
For bridges in Seismic Zone 2, the design
displacements are specified as the maximum of those
determined from the elastic analysis of A4.7.4.3 and
D4.7.4.3 or the minimum specified support lengths given by
Equation A4.7.4.4-1.
This either/or specification was
introduced to account for larger displacements that may
occur from the analysis of more flexible bridges. It was the
opinion of the PEP that displacement obtained from the
elastic analysis of bridges should provide a reasonable
N
=
minimum support length measured perpendicular
to abutment or pier face from the end of the beam
at the centerline of the bottom flange.
The N calculated in Equation A1 shall not be taken less
than 300 mm {12 in.}.
B.4 - 37
September 2007
SPECIFICATIONS
COMMENTARY
NOTE: S in Equation A1 is based on AASHTO's
definition of skew angle, see PP3.2.2 for PennDOT's and
AASHTO's definition of skew angle.
estimate of the displacements resulting from the inelastic
response of the bridge. However, it must be recognized that
displacements are very sensitive to the flexibility of the
foundation; if the foundation is not included in the elastic
analysis of A4.7.4.3 and D4.7.4.3, consideration should be
given to increasing the specified displacements for bridges
founded on very soft soils. This increase may be of the
order of 50 percent or more, but, as with any generalization,
considerable judgment is required. A better method is to
determine upper and lower bounds from an elastic analysis
which incorporates foundation flexibility. Special care in
regard to foundation flexibility is required for bridges with
high piers.
Table 4.7.4.4-1 – Percentage N by Zone and
Acceleration Coefficient
ZONE
ACCELERATION
COEFFICIENT
SOIL
TYPE
%N
1
0.025
I or II
100
1
0.025
III or IV
100
1
≥0.025
All
100
2
All Applicable
All
100
3
All Applicable
All
100
4
All Applicable
All
100
Figure C4.7.4.4-1 - Dimensions for Minimum Support
Length Requirements
4.7.4.5P BASE ISOLATION DESIGN
C4.7.4.5.1P
4.7.4.5.1P General
This article incorporates generic requirements for
seismic isolation design. The isolation of structures from
the damaging effects of earthquakes is not a new idea. The
first patents for base isolation schemes were taken out at the
turn of the century, but, until very recently, few structures
were built which use these ideas. Early concerns were
focused on the fear of uncontrolled displacements at the
isolation interface, but these have been largely overcome by
the successful development of mechanical energy
dissipators. When used in combination with a flexible
device, such as an elastomeric bearing or a sliding plate, an
energy dissipator can control the response of an isolated
structure by limiting both the displacements and the forces.
Interest in base isolation, as an effective means of protecting
bridges from earthquakes, has, therefore, revived in recent
years. To date, there are several hundred bridges in New
Zealand, Japan, Italy, and the United States which use base
isolation principles and technology in their seismic design.
The intent of seismic isolation is to increase the
fundamental period of vibration so that the structure is
subjected to lower earthquake forces.
However, the
This article includes the fundamental requirements for
base isolation design, which greatly reduces the earthquake
forces that a bridge must resist.
The same acceleration coefficient, A, is prescribed for
base isolation design as for cases without base isolation.
This coefficient is given on a county-by-county basis in
Figure D3.10.2-1. However, a minimum value of 0.1 shall
be used.
Seismic zone is not a delineator concerning method of
analysis; however, minimum design requirements are
governed by the seismic zone. Refer to Table A3.10.4-1 for
the appropriate seismic zone designation.
The site coefficient for base isolation design, Si, which
accounts for the effects of the site condition on the elastic
response coefficient, is given in Table 1. The soil profile
types are the same as those described in A3.10.5.
The R-factors presented in Table A3.10.7.1-1 and
A3.10.7.1-2 are used for base isolation design with the
exception that the R-factor for all substructures shall not be
B.4 - 38
September 2007
SPECIFICATIONS
COMMENTARY
greater than 2.0.
Each statically stable segment of the structure shall be
analyzed for the statically equivalent seismic force given in
D4.7.4.5.2P. For Seismic Zone 2, the requirements of
D4.7.4.5.6P shall apply.
reduction in force is accompanied by an increase in
displacement demand which must be accommodated within
the flexible mount. Furthermore, longer period bridges can
be lively under service loads.
There are, therefore, three basic elements in those base
isolation systems that have been used to date. These are:
Table 4.7.4.5.1P-1 – Site Coefficient for Base Isolation (Si)
(a) A flexible mounting so that the period of vibration of
the total system is lengthened sufficiently to reduce the
force response.
Soil Profile Type
Site
Coefficient
I
II
III
IV
IV
Si
1.0
1.5
2.0
3.0
3.0
(b) A damper or energy dissipater so that the relative
deflections across the flexible mounting can be limited
to a practical design level.
(c) A means of providing rigidity under low (service) load
levels such as wind and braking forces.
Flexibility
An elastomeric bearing is not the only means of
introducing flexibility into a structure, but it certainly
appears to be the most practical and the one with the widest
range of application. The idealized force response with
increasing period (flexibility) is shown schematically in the
force response curve in Figure C1. Reductions in base shear
occur as the period of vibration of the structure is
lengthened. The extent to which these forces are reduced is
primarily dependent on the nature of the earthquake ground
motion and the period of the fixed base structure. However,
as noted above, the additional flexibility needed to lengthen
the period of the structure will give rise to large relative
displacements across the flexible mount. Figure C2 shows
an idealized displacement response curve from which
displacements are seen to increase with increasing period
(flexibility).
Figure C4.7.4.5.1P-1 - Idealized Force Response Curve
B.4 - 39
SPECIFICATIONS
September 2007
COMMENTARY
Figure C4.7.4.5.1P-2 - Idealized Displacement Response
Curve
Energy Dissipation
Large relative displacements can be controlled if
substantial additional damping is introduced into the
structure at the isolation level. This is shown schematically
in Figure C3. It can also be seen that higher damping
removes much of the sensitivity to variations in ground
motion characteristics, as is indicated by the smoother force
response curves at higher damping.
One of the most effective means of providing a
substantial level of damping is through hysteretic energy
dissipation. The term "hysteretic" refers to the offset
between the loading and unloading curves under cyclic
loading. Figure C4 shows an idealized force-displacement
loop where the enclosed area is a measure of the energy
dissipated during one cycle of motion. Mechanical devices
which use the plastic deformation of either mild steel or lead
to achieve this behavior have been developed.
B.4 - 40
SPECIFICATIONS
September 2007
COMMENTARY
Figure C4.7.4.5.1P-3 - Response Curves for Increasing
Damping
Figure C4.7.4.5.1P-4 - Idealized Hysteresis Loop
Rigidity Under Low Lateral Loads
While lateral flexibility is highly desirable for high
seismic loads, it is clearly undesirable to have a structural
system that will vibrate perceptibly under frequently
occurring loads such as wind loads or braking loads.
Mechanical energy dissipators may be used to provide
rigidity at these service loads by virtue of their high initial
elastic stiffness. Alternately, some base isolation systems
require a separate wind restraint device for this purpose typically a rigid component which is designed to fail at a
B.4 - 41
SPECIFICATIONS
September 2007
COMMENTARY
given level of lateral load.
Design Application
The seismic design principles fodr base isolation are
best illustrated by Figure C5 which was based on the
AASHTO Standard Specification for Highway Bridges..
The same basic concept concerning base isolation also
applies to the LRFD Specification. The solid uppermost
line is the realistic (elastic) ground response spectrum as
recommended in the AASHTO Standard Specification for
Highway Bridges for the highest seismic zone. This is the
spectrum that is used to determine actual forces and
displacements for conventional design. The lowest solid
line is the design curve from the AASHTO Standard
Specification for Highway Bridges. It is seen to be
approximately one-fifth of the realistic forces given by the
AASHTO Standard Specification for Highway Bridges.
This reduction, to obtain the design forces, is consistent with
an R-factor of 5 for a multi-column bent. Also shown is the
probable overstrength of a bent so designed. Now if the
bridge is isolated, the shear forces in this multi-column bent
may be represented by an inelastic spectrum that
incorporates the damping of the isolation system, as shown
by the small dashed line in Figure C5.
Figure C4.7.4.5.1P-5 – Earthquake Forces
B.4 - 42
September 2007
SPECIFICATIONS
COMMENTARY
The period of the bridge will be in the 2.0- to 2.5second range; in this range the overstrength (actual capacity)
of the bent exceeds the realistic forces (demand) for the
isolated bridge. This region has been shaded in Figure C5.
There is, therefore, no inelastic deformation or ductility
required of the bent, and elastic performance (without
damage) is ensured. The benefits of seismic isolation for
bridges may be summarized as follows:
(a) Reduction in the realistic forces to which a bridge will
be subjected by a factor of between 5 and 10.
(b) Elimination of the ductility demand and, hence, damage
to the piers.
(c) Control of the distribution of the seismic forces to the
substructure elements with appropriate sizing of the
elastometric bearings.
(d) Reduction in column design forces by a factor of
approximately 2 in comparison with conventional
design.
(e) Reduction in foundation design forces by a factor of
between 2 and 3 in comparison with conventional
design.
The intent of seismic isolation design is to eliminate or
significantly reduce damage (inelastic deformation) to the
substructure. By limiting the R-factor to a maximum value
of 2, the overstrength inherent in the substructures will
ensure minimal ductility demand on the substructure for the
design earthquake. For essential structures, consideration
may be given to reducing the maximum value of R to as low
as 1 to ensure complete elastic response of the substructure.
4.7.4.5.2P Statically Equivalent Seismic Force and
Coefficient
C4.7.4.5.2P
The statically equivalent force is given by:
F=CsW
(4.7.4.5.2P-1)
where:
CS =
elastic seismic response coefficient
W =
weight of the superstructure segment supported by
isolation bearings
For seismic isolation design, the elastic seismic
coefficient is again related to the elastic ground response
spectra. Here, the form is slightly different from that for
non-isolation design (which involved T2/3) and, for 5%
damping, is given by
Cs
The elastic seismic response coefficient, CS, used to
determine the equivalent force is given by the dimensionless
relation:
ASi
T
(C4.7.4.5.2P-1)
In this case, A is once again the acceleration
coefficient, Si is the site coefficient for base isolation (Table
D4.7.4.5.1P-1), and the 1/T factor accounts for the decrease
in the response spectra ordinates as T increases. The
specific Si values reflect the fact that above a period of 1.0
second, there is a 1.0 to 1.5 to 2.0 to 3.0 relationship for the
spectral accelerations for soil profile Types I, II, III and IV,
B.4 - 43
September 2007
SPECIFICATIONS
COMMENTARY
Kbdi
W
where:
Cs
Kb =
W =
di
=
(4.7.4.5.2P-2)
equivalent linear stiffness of all
supporting the superstructure segment
bearings
CS
weight of the superstructure segment supported on
the isolation bearings
displacement across the isolation bearings
The displacement di (mm) {in.} is given by
Metric Units: d i
250 ASiT
B
U.S. Customary Units: d i
respectively. Once again, Cs should not exceed a value of
2.5A.
If the effects of damping are included, the elastic
seismic coefficient is given by
(4.7.4.5.2P-3)
ASi
TB
(C4.7.4.5.2P-2)
where B is the damping term given in Table D4.7.4.5.2P-1.
Note that for 5% damping, B = 1.0.
The quantity Cs is a dimensionless design coefficient
which, when multiplied by g, produces the spectral
acceleration. This spectral acceleration, SA, is related to the
spectral displacement, SD, by the relationship
SA
2
(C4.7.4.5.2P-3)
SD
Where ω is the circular natural frequency and is given by
2 π/ T. Therefore, since SA ≈ Csg, then
10 ASiT
B
where:
ASi
g
TB
A
=
acceleration coefficient as defined in D4.7.4.5.1P
SA
B
=
damping factor given in Table D4.7.4.5.2P-1
and
Si
=
dimensionless site coefficient for isolation design
for the given soil profile as designated in
D4.7.4.5.1P
Metric Units:
T
T
=
2
period of vibration (seconds) given by
W
Kb g
=
1
2
T2
2
(4.7.4.5.2P-4)
where:
g
SD
(C4.7.4.5.2P-4)
ASi
g
TB
(C4.7.4.5.2P-5)
ASi
9807 mm / s 2
TB
248.4 ASiT
mm
B
acceleration of gravity (mm/s2) {in/sec2}
SD
1
2
T2
2
ASi
TB
ASi
386.4in / sec 2
TB
9.79 ASiT
in.
B
Denoting SD as di, which is the displacement across the
elastomeric bearings, the above is approximated to be
Metric Units: d i
B.4 - 44
250 ASiT
mm
B
(C4.7.4.5.2P-6)
September 2007
SPECIFICATIONS
COMMENTARY
U.S. Customary Units: d i
10 ASiT
in
B
An alternate form for Cs is possible. The quantity Cs is
defined by the relationship
F
(C4.7.4.5.2P-7)
CSW
where:
F = earthquake design force
W = weight of the structure
Therefore,
Cs
Kbdi
W
F
W
where Kb is the equivalent linear spring of all bearings
supporting the superstructure segment (see Figure
DC4.7.4.5.1P-4). The equivalence of this form to the
previous form is evident by recalling that Kb = ω 2 W/g,
from which
Metric Units:
2
W
g
CS
di
W
2 2
T
1
9807
248.4 ASiT
B
ASi
BT
(C4.7.4.5.2P-9)
2
W
g
CS
Table 4.7.4.5.2P-1 – Damping Coefficient B
Damping
(% of Critical)*
≤2
5
10
20
30
40
>50
B
0.8
1.0
1.2
1.5
1.7
1.9
2.0
*The percent of critical damping shall be verified by test of the
isolation system’s characteristics. The damping coefficient shall be
based on linear interpolation for damping levels other than those given.
4.7.4.5.3P Requirements for Elastic Force Determination
The statically equivalent force determined according to
D4.7.4.5.2P, which is associated with the displacement
across the isolation bearings, shall be applied independently
along two perpendicular axes and combined as specified in
B.4 - 45
di
W
2 2
T
1
386.4
9.79 ASiT
B
ASi
BT
September 2007
SPECIFICATIONS
COMMENTARY
A3.10.8.
Typically, the perpendicular axes are the longitudinal
and transverse axes of the bridge, but the choice is open to
the designer. The longitudinal axis of a curved bridge may
be a chord connecting the two abutments.
4.7.4.5.4P Design Displacement for Other Loads
C4.7.4.5.4P
Adequate clearance must be provided to permit shear
deflections in the bearings resulting from braking loads,
wind loads and centrifugal forces. These deflections are a
function of the force-deflection characteristics of the
bearings.
To protect against fatigue under frequently occurring
loads, deflections should be kept as small as practicable. As
a guide, the deflection due to braking load should be less
than 10% of the deformable rubber thickness, and that for
extreme wind loads should be less than 25% of the
deformable rubber thickness.
4.7.4.5.5P Design Forces for Seismic Zone 1
C4.7.4.5.5P
The design force for the connection between
superstructure and substructure at each bearing is given by
Seismic isolation design provides a significant
reduction in the real elastic forces, and the exception of this
section permits this force reduction to be utilized in the
design of the bearings. However, it should be noted that the
acceleration coefficient, which has a maximum value of
0.09 for bridges in Seismic Zone 1, is specified to have a
minimum value of 0.10 if seismic isolation is used. This
conservatism will ensure, for most areas in Seismic Zone 1,
that the isolation bearings are capable of resisting twice the
design earthquake. This level of conservatism is greater
than that inherent in the requirement for bearings being
designed for 0.2 times the dead load.
F A = K bd i
(4.7.4.5.5P-1)
where:
Kb =
equivalent linear stiffness of the isolation bearing
di
displacement of the isolated bridge deck as
specified in D4.7.4.5.2P, using a minimum
acceleration coefficient, A, of 0.10
=
4.7.4.5.6P Design Forces for Seismic Zone 2
The requirements of D4.7.4.5.2P shall apply, except
that in the determination of seismic design forces an Rfactor of 1 shall be used.
4.7.4.5.7P Substructure Design Requirements
4.7.4.5.7.1P Foundations and Abutments
The provisions of this document and the LRFD shall
apply to the design of foundations and abutments.
4.7.4.5.7.2P Columns, Footings and Connections
4.7.4.5.7.2aP Structural Steel
The provisions of this document and the LRFD shall
apply to the design of structural steel columns and
connections.
4.7.4.5.7.2bP Reinforced Concrete
The provision of this document and the LRFD shall
apply to the design of reinforced concrete columns, pier
footings and connections.
B.4 - 46
September 2007
SPECIFICATIONS
COMMENTARY
4.8 ANALYSIS BY PHYSICAL MODELS
4.8.2 Bridge Testing
When approved by the Chief Bridge Engineer, existing
bridges may be instrumented and results obtained under
various conditions of traffic and/or environmental loads or
load tested with special purpose vehicles to establish force
effects and/or the load carrying capacity of the bridge.
B.4 - 47
DM-4, Section 6 - Steel Structures
September 2007
DESIGN MANUAL
PART 4
VOLUME 1
SECTION 6 – STEEL STRUCTURES
6.1 Scope .................................................................................................................................................................................... 1
6.1.1P Restrictions of Steel Bridge Types ........................................................................................................................ 1
6.1.1.1P Steel Tied-Arch Bridges ................................................................................................................................. 1
6.2 Definitions .................................................................................................................................................................... 2
6.3 Notation ............................................................................................................................................................................... 2
6.4 Materials .............................................................................................................................................................................. 3
6.4.1 Structural Steels ........................................................................................................................................................ 3
6.4.3 Bolts, Nuts and Washers .......................................................................................................................................... 4
6.4.3.1 Bolts.................................................................................................................................................................. 4
6.4.3.2 Nuts .................................................................................................................................................................. 5
6.4.3.3 Washers ............................................................................................................................................................ 5
6.4.3.4 Alternative Fasteners ........................................................................................................................................ 5
6.4.3.5 Load Indicator Devices ..................................................................................................................................... 5
6.4.7 Stainless Steel ............................................................................................................................................................ 6
6.5 Limit States.......................................................................................................................................................................... 6
6.5.2 Service Limit State .................................................................................................................................................... 6
6.5.3 Fatigue and Fracture Limit State ............................................................................................................................ 6
6.5.4 Strength Limit State ................................................................................................................................................. 6
6.5.4.2 Resistance Factors ............................................................................................................................................ 6
6.6 Fatigue and Fracture Considerations ................................................................................................................................ 7
6.6.1 Fatigue ....................................................................................................................................................................... 7
6.6.1.2 Load-Induced Fatigue ....................................................................................................................................... 7
6.6.1.2.1 Application .............................................................................................................................................. 7
6.6.1.2.2 Design Criteria ....................................................................................................................................... 7
6.6.1.2.4 Restricted Use Details ............................................................................................................................ 8
6.6.1.2.5 Fatigue Resistance .................................................................................................................................. 8
6.6.1.3 Distortion-Induced Fatigue ............................................................................................................................... 8
6.6.1.3.4P Distortion-Induced Fatigue: Unacceptable Details and Acceptable Alternative Details ..................... 9
6.6.2 Fracture ................................................................................................................................................................... 16
6.7 General Dimension and Detail Requirements ................................................................................................................ 16
6.7.2 Dead Load Camber................................................................................................................................................. 16
6.7.2.1P Camber Due to Weight of Deck Slab ........................................................................................................... 16
6.7.2.2P Camber Details for Design Drawings ........................................................................................................... 16
6.7.2.3P Heat-Curve Camber Corrections .................................................................................................................. 18
6.7.3 Minimum Thickness of Steel .................................................................................................................................. 19
6.7.4 Diaphragms and Cross-Frames ............................................................................................................................. 19
6.7.4.1 General ........................................................................................................................................................... 19
6.7.4.2 I-Section MEMBERS ..................................................................................................................................... 20
6.7.4.3 Box Section MEMBERS ................................................................................................................................ 21
6.7.5 Lateral Bracing ....................................................................................................................................................... 23
6.7.5.2 I-Section MEMBERS ..................................................................................................................................... 23
6.7.5.3 Tub Section MEMBERs ................................................................................................................................. 23
6.8 Tension Members .............................................................................................................................................................. 24
6.8.2 Tensile Resistance ................................................................................................................................................... 24
6.8.2.2 Reduction Factor, U ........................................................................................................................................ 24
6.8.2.3 Combined Tension and Flexure ...................................................................................................................... 24
B.6 - lxiv
September 2007
6.8.3 Net Area ................................................................................................................................................................... 25
6.9 Compression Members ..................................................................................................................................................... 25
6.9.2 Compressive Resistance ......................................................................................................................................... 25
6.9.2.2 Combined Axial Compression and Flexure .................................................................................................... 25
6.9.5 Composite Members ............................................................................................................................................... 25
6.10 I-Sections In Flexure ....................................................................................................................................................... 25
6.10.0P Applicable Provisions .......................................................................................................................................... 25
6.10.1 General .................................................................................................................................................................. 26
6.10.1.1.1 Stresses ................................................................................................................................................ 27
6.10.1.1.1d Concrete Deck Stresses ............................................................................................................. 28
6.10.1.1.1.fP Lateral Support of Top Flanges Supporting Timber Decks .................................................... 28
6.10.1.2 Noncomposite Sections ................................................................................................................................. 28
6.10.1.3 Hybrid Sections ............................................................................................................................................ 29
6.10.1.4 Variable Web Depth Members ..................................................................................................................... 29
6.10.1.5 Stiffness ......................................................................................................................................................... 29
6.10.1.6 Flange Stresses And Member Bending Moment .......................................................................................... 29
6.10.1.7 Minimum Negative Flexure Concrete Deck Reinforcement ........................................................................ 31
6.10.1.11p Lateral Support Of Top Flanges Supporting Timber Decks ..................................................................... 32
6.10.3 Constructibility ...................................................................................................................................................... 32
6.10.3.2 Flexure ........................................................................................................................................................... 32
6.10.3.2.1 Discretely Braced Flanges in Compression ......................................................................................... 32
6.10.3.2.5.1P Slab Placement .............................................................................................................................. 32
6.10.3.2.5.2P Deck Slab Overhang Form Support .............................................................................................. 33
6.10.3.2.4.3P Deck Slab Overhang Rotation ....................................................................................................... 37
6.10.3.5 Dead Load Deflection .................................................................................................................................... 37
6.10.4 Service Limit State ................................................................................................................................................. 37
6.10.4.2 Permanent Deformation ................................................................................................................................. 37
6.10.4.2.1 General ................................................................................................................................................ 37
C6.10.4.2.1 ........................................................................................................................................................... 37
6.10.4.2.2 Flexure ................................................................................................................................................. 37
6.10.5 Fatigue and Fracture Limit State ......................................................................................................................... 38
6.10.5.1 Fatigue ........................................................................................................................................................... 38
6.10.5.3 Special Fatigue Requirement for Webs ........................................................................................................ 38
6.10.6 Strength Limit State .............................................................................................................................................. 38
6.10.6.2 Flexure ........................................................................................................................................................... 38
6.10.6.2.2 Composite Sections in Positive Flexure .............................................................................................. 38
6.10.6.2.3 Composite Sections in Negative Flexure and Noncomposite Sections ............................................... 38
6.10.8 Flexural Resistance - Composite Sections in Negative Flexure and Noncomposite Sections ........................ 39
6.10.8.2 Compression-Flange Flexural Resistance ...................................................................................................... 39
6.10.8.2.3 Lateral Torsional Buckling Resistance ............................................................................................... 39
6.10.9 Shear Resistance ................................................................................................................................................... 40
6.10.9.1 General ......................................................................................................................................................... 40
6.10.9.3.3 End Panels .......................................................................................................................................... 40
6.10.10 Shear Connectors ................................................................................................................................................ 40
6.10.10.1 General ....................................................................................................................................................... 40
6.10.10.1.1 Types ................................................................................................................................................ 40
6.10.10.1.2 Pitch .................................................................................................................................................. 40
6.10.10.1.3 Transverse Spacing ........................................................................................................................... 42
6.10.10.1.4 Cover and Penetration ...................................................................................................................... 42
6.10.10.1.5p Splice Locations .................................................................................................................................... 42
6.10.10.2 Fatigue Resistance ...................................................................................................................................... 42
6.10.10.3 Special Requirements for Points of Permanent Load Contraflexure .......................................................... 43
6.10.10.4 Strength Limit State .................................................................................................................................... 43
6.10.10.4.2 Nominal Shear Force ........................................................................................................................ 43
6.10.11 Stiffeners .............................................................................................................................................................. 45
6.10.11.1 Transverse Intermediate Stiffeners ............................................................................................................. 45
6.10.11.1.1 General .............................................................................................................................................. 45
6.10.11.1.2 Projecting Width ............................................................................................................................... 45
B.6 - lxv
September 2007
6.10.11.3 Longitudinal Stiffeners ............................................................................................................................... 45
6.10.11.3.3 Moment of Inertia and Radius of Gyration ....................................................................................... 46
6.10.11.4P Stiffeners in Rigid-Frame Knees .............................................................................................................. 47
6.10.11.4.1P Stiffener Spacing ............................................................................................................................ 47
6.10.11.4.2P Stiffener Design ............................................................................................................................. 48
6.10.12 Cover Plates......................................................................................................................................................... 49
6.10.12.3P Cover Plate Length and Width ................................................................................................................. 49
6.11 Box Sections in Flexure ................................................................................................................................................... 49
6.11.1 General ................................................................................................................................................................... 49
6.11.1.1 Stress Determination...................................................................................................................................... 50
6.11.3 Constructibility ...................................................................................................................................................... 50
6.11.3.2 Flexure ........................................................................................................................................................... 50
6.11.5 Fatigue and Fracture Limit State ......................................................................................................................... 50
6.11.6 Strength Limit State ............................................................................................................................................. 51
6.11.6.2 Flexure .......................................................................................................................................................... 51
6.11.6.2.2 Sections in Positive Flexure ............................................................................................................... 51
6.11.9 Shear Resistance .................................................................................................................................................... 51
6.11.10 Shear Connectors ................................................................................................................................................. 51
6.11.11 Stiffeners .............................................................................................................................................................. 52
6.11.11.2 Longitudinal Compression-Flange Stiffeners ............................................................................................. 52
6.12 Miscellaneous Flexural Members .................................................................................................................................. 52
6.12.1 General .................................................................................................................................................................. 52
6.12.1.2 Strength Limit State ...................................................................................................................................... 52
6.12.1.2.3 Shear ................................................................................................................................................... 52
6.12.2 Nominal Flexural Resistance ............................................................................................................................... 52
6.12.2.2 Noncomposite Members ............................................................................................................................... 52
6.12.2.2.1 I- and H-Shaped Members.................................................................................................................. 52
6.13 Connections and Splices ................................................................................................................................................. 53
6.13.1 General .................................................................................................................................................................. 53
6.13.2 Bolted Connections ............................................................................................................................................... 53
6.13.2.1 General ......................................................................................................................................................... 53
6.13.2.3 Bolts, Nuts and Washers ............................................................................................................................... 54
6.13.2.3.2 Washers .............................................................................................................................................. 54
6.13.2.4 Holes ............................................................................................................................................................. 54
6.13.2.4.1 Types .................................................................................................................................................. 54
6.13.2.4.1b Oversize Holes........................................................................................................................... 54
6.13.2.4.2 Size ..................................................................................................................................................... 54
6.13.2.5 Size of Bolts ................................................................................................................................................. 55
6.13.2.6 Spacing of Bolts ........................................................................................................................................... 56
6.13.2.6.6 Edge Distances ................................................................................................................................... 56
6.13.2.7 Shear Resistance ........................................................................................................................................... 57
6.13.2.8 Slip Resistance .............................................................................................................................................. 57
6.13.2.10 Tensile Resistance ............................................................................................................................... 58
6.13.2.10.3 Fatigue Resistance ............................................................................................................................ 58
6.13.2.11 Combined Tension and Shear ..................................................................................................................... 59
6.13.3 Welded Connections ............................................................................................................................................. 59
6.13.3.1 General ......................................................................................................................................................... 59
6.13.3.8p Intersecting Welds ...................................................................................................................................... 59
6.13.3.9P Intermittent Fillet Welds ............................................................................................................................ 60
6.13.3.10P Minimum Edge Distance .......................................................................................................................... 60
6.13.4 Block Shear Rupture Resistance ......................................................................................................................... 60
6.13.6 Splices .................................................................................................................................................................... 60
6.13.6.1 Bolted Splices ............................................................................................................................................... 60
6.13.6.1.1 General ............................................................................................................................................... 60
6.13.6.1.4 Flexural Members ............................................................................................................................... 60
6.13.6.1.4a General ....................................................................................................................................... 61
6.13.6.1.4b Web Splices ............................................................................................................................... 61
6.13.6.1.4c Flange Splices ........................................................................................................................... 62
B.6 - lxvi
September 2007
6.13.6.1.5 Fillers .................................................................................................................................................. 63
6.13.6.2 Welded Splices ............................................................................................................................................. 63
6.15 Piles .................................................................................................................................................................................. 63
6.15.1 General .................................................................................................................................................................. 63
6.15.2 Structural Resistance............................................................................................................................................ 65
6.15.3P Compressive Resistance ..................................................................................................................................... 66
6.15.3.1 Axial Compression ....................................................................................................................................... 66
6.15.3.2 Combined Axial Compression and Flexure .................................................................................................. 67
6.15.3.3 Buckling ....................................................................................................................................................... 70
6.15.4 Maximum Permissible Driving Stresses ............................................................................................................. 70
Appendix A – Flexural Resistance – Composite Sections in Negative Flexure and Noncomposite Sections with
Compact or Noncompact Webs ....................................................................................................................................... 71
Appendix B – Moment Redistribution from Interior-Pier Sections in Continuous-Span Bridges ................................... 71
Appendix C – Basic Steps for Steel Bridge Superstructures ................................................................................................ 71
B.6 - lxvii
September 2007
SPECIFICATIONS
COMMENTARY
6.1 SCOPE
6.1.1P Restrictions of Steel Bridge Types
6.1.1.1P STEEL TIED-ARCH BRIDGES
C6.1.1.1P
Steel tied-arch bridges shall be used only after thorough
consideration has been given to all factors in design,
fabrication and erection, and if the design is approved by the
Chief Bridge Engineer. In the preliminary stage, the tiedarch must show a marked economic advantage over
alternate designs to warrant further consideration. Tied-arch
structures are currently unacceptable to FHWA. Moreover,
refer to the FHWA Technical Advisory T-5140.4, dated
September 28, 1978, for the problems pertinent to tied-arch
structures.
Transverse welds on the tie girders shall be avoided, where
possible. Bolted connections shall be used instead of
transverse welds.
On Langer-girder tied-arch bridges (those tied arches where
the tie girder acts as the major flexural member in addition
to providing horizontal reactions to the arch rib) with box
girders functioning as tie girders, the internal diaphragms
stiffening the box at the floorbeam connections shall be
attached to both flanges, as well as the webs. A tie plate
should be placed between the tie-girder flange and the
floorbeam flange if they lie essentially in the same plane.
Hangers composed of multiple bridge strands shall have
either spacers between the strands or dampers, or both.
The dynamic response of the bridge due to traffic shall be
investigated by an appropriate three-dimensional, forcedvibration dynamic analysis, especially for tied-arch bridges
that do not employ Langer-girders.
Steel tied-arch bridges have experienced such problems as
lamellar tearing in the hanger connections, detrimental
vibration in the main structure and cables, and cracking in
fracture-critical members. The design, detailing, and
fabrication of the floorbeams are critical for long-term
performance. Fatigue cracking has occurred in floorbeams
due to out-of-plane distortion in combination with abrupt
termination of the flange; proper coping and grinding o fthe
cope were not performed.
6.1.1.2P Steel Box Bridges
C6.1.1.2P
Steel box bridges shall be used only after thorough
consideration has been given to all factors in design,
fabrication, erection and future in-depth inspection, and if
the design is approved by the Chief Bridge Engineer. In the
preliminary stage, the steel box design must show a marked
economic or aesthetic advantage over alternate designs to
warrant further consideration.
Even though steel box girders may provide aesthetically
pleasing and sometimes economical structures, the
Department has major concerns about steel box girders
which are:
The designer must use intuitive engineering judgment when
selecting the type, location and number of spacers used
between the strands of a hanger composed of multiple
bridge strands. The need for spacers is not based upon a
calculated analysis, but rather on the observation that some
bridges without spacers experienced problems and were
subsequently retrofitted with spacers.
difficult inspection environment,
inspection complexities,
future cleaning, painting and/or repair difficulties.
detailing complexities
stability during erection
B.6 - 1
SPECIFICATIONS
September 2007
COMMENTARY
6.2 Definitions
Controlling Flange—top or bottom flange for the smaller section at a point of splice, whichever flange has the maximum
elastic flexural stress at its mid-thickness due to the factored loads.
Non-controlling Flange—the flange at a point of splice opposite the controlling flange.
6.3 NOTATION
The following shall supplement A6.3
Abot
Ad
=
=
Ffat
=
Ffat1
=
Ffat2
=
Fp
=
Frc
FT
=
=
fbu
=
Ln
=
Lp
=
Mu
=
PT
=
R
r
=
=
Vfat
w
Z
=
=
=
=
=
flg
area of the bottom flange (mm2) {in.2) (D6.10.10.1.2)
minimum required cross-sectional area of a diagonal member of top lateral bracing for tub sections (mm 2) {in.2)
(DC6.7.5.3)
radial fatigue shear range per unit length, taken as the larger of either Ffat1 or Ffat2 (N/mm) {kip/in.)
(D6.10.10.1.2)
radial fatigue shear range per unit length due to the effect of any curvature between brace points (N/mm)
{kip/in.) (D6.10.10.1.2)
radial fatigue shear range per unit length due to torsion caused by effects other than curvature, such as skew
(N/mm) {kip/in.) (D6.10.10.1.2)
total radial shear force in the concrete deck at the point of maximum positive live load plus impact moment for
the design of the shear connectors at the strength limit state, taken equal to zero for straight spans or segments
(N) {kip) (D6.10.10.4.2)
net range of cross-frame force at the top flange (N) {kip) (D6.10.10.1.2)
total radial shear force in the concrete deck between the point of maximum positive live load plus impact
moment and the centerline of an adjacent interior support for the design of shear connectors at the strength limit
state, taken equal to zero for straight spans or segments (N) {kip) (D6.10.10.4.2)
largest value of the compressive stress throughout the unbraced length in the flange under consideration,
calculated without consideration of flange lateral bending (MPa) {ksi) (D6.10.1.6)
arc length between the point of maximum positive live load plus impact moment and the centerline of an
adjacent interior support (mm) {ft.) (D6.10.10.4.2)
arc length between an end of the girder and an adjacent point of maximum positive live load plus impact
moment (mm) {ft.) (D6.10.10.4.2)
largest value of the major-axis bending moment throughout the unbraced length causing compression in the
flange under consideration (N-mm) {k-in.) (D6.10.1.6)
total longitudinal shear force in the concrete deck between the point of maximum positive live load plus impact
moment and the centerline of an adjacent interior support for the design of the shear connectors at the strength
limit state, taken as the sum of Pp and Pn (N){kip) (D6.10.10.4.2)
minimum girder radius within a panel (mm) {ft.) (D6.7.4.2)
desired bending stress ratio in a horizontally curved I-girder, taken equal to f fbu (DC6.7.4.2)
longitudinal fatigue shear range per unit length (N/mm) {kip/in.) (D6.10.10.1.2)
effective length of deck assumed acting radial to the girder (mm) {in.) (D6.10.10.1.2)
curvature parameter for determining required longitudinal web stiffener rigidity (D6.10.11.3.3)
curvature correction factor for longitudinal web stiffener rigidity (D6.10.11.3.3)
range of longitudinal fatigue stress in the bottom flange without consideration of flange lateral bending (MPa)
{ksi} (D6.10.10.1.2)
B.6 - 2
September 2007
SPECIFICATIONS
COMMENTARY
6.4 MATERIALS
6.4.1 Structural Steels
C6.4.1
Poisson's ratio for structural steel shall be assumed to be 0.3
in the elastic range.
Unless directed otherwise, all structural steel shall conform
to the specifications for structural steel, ASTM
A 709/A 709M, Grade 250 {Grade 36}. Other types of
steel, such as ASTM A 709/A 709M, Grades 345 and 345W
{Grade 50 and 50W}, in combination with ASTM
A 709/A 709M, Grade 250 {Grade 36}, or with each other
may be considered for economy.
Steel grades 690 or 690W {Grades 100 or 100W} shall not
be used unless written approval has been obtained from the
Chief Bridge Engineer.
Unpainted ASTM A 709/A 709M, Grade 345W or HPS485W {Grade 50W or HPS-70W}, steel shall not be
specified without written approval of the Chief Bridge
Engineer at the TS&L stage. This policy applies to state
and local bridges and bridges where State or Federal
funding is utilized. Use in contractor-designed alternates
must also be approved at the TS&L stage. Use is not
permitted in acidic or corrosive environments, in locations
subject to salt water spray or fog, in depressed roadway
sections (less than 6100 mm {20 ft.} clearance) where salt
spray and other pollutants may be trapped, in low
underclearance situations where the steel is either less than
1500 mm {5 ft.} from normal water elevation or
continuously wet, or where the steel may be buried in soil.
The use of Grade 345W or HPS-485W {Grade 50W or
HPS-70W} steel is not permitted in bridge types where salt
spray and dirt accumulation may be a concern (e.g., trusses
or inclined-leg bridges) unless corrosion-susceptible regions
are painted.
Do not use Grade 345W or HPS-485W {Grade 50W or
HPS-70W} steel for expansion dams, or for stringers or
other members under open steel decking.
Where the use of Grade 345W {Grade 50W} or Grade HPS485W {Grade HPS-70W}unpainted weathering steel is
permitted, the following criteria must be met:
For additional information on the economics of steel
bridges, see PP4.3.
For additional information, refer to NCHRP Report No. 314,
Guidelines for the Use of Weathering Steel in Bridges.
(a) The number of expansion joints shall be
minimized.
(b) Details to avoid retention of water and debris shall
be incorporated in the design.
(c) The steel shall be painted to a length of at least 1.5
times web depth and a minimum of 1500 mm {5
ft.} on each side of the expansion joint.
(d) Drip plates shall be provided.
Drip bars attached as indicated on BC-753M.
B.6 - 3
September 2007
SPECIFICATIONS
COMMENTARY
(e) The substructure units shall be protected against
staining. Use special drainage details for pier and
abutment tops and/or protective coating for
reinforced concrete surfaces in accordance with the
Publication 408.
(f) Mechanical fasteners made of ASTM A 325 and
A 490, Type 3, weathering steels and stainless
steels are suitable for weathering steel bridges. Do
not use zinc and cadmium galvanized carbon-steel
bolts for weathering steel bridges.
Preferably for weathering steel bridges, use mechanical
fasteners made of weathering steel. When stainless steel
mechanical fasteners are used with weathering steel bridges,
there is a possibility of galvanic corrosion of the weathering
steel. Due to the small area of the bolt in relation to the
material being bolted, the effect is usually negligible.
(g) Load indicator washers are not recommended.
For existing bridges, where Grade 345W {Grade 50W}
unpainted steel is used, clean and paint the beam ends up to
1500 mm {5 ft.} from leaking joints, or to where the
weathering steel area is exposed to or subject to salt water
spray.
6.4.3 Bolts, Nuts and Washers
6.4.3.1 BOLTS
C6.4.3.1
Bolts shall conform to one of the following:
Although there are metric high-strength bolt standards for
ASTM A 325M and A 490M, as of this writing, no
installation specification is available for metric highstrength bolts and no domestic high-strength bolt
manufacturers are producing the metric bolts. The
Department decided to use soft metric conversions of
standard "English" high-strength bolts (ASTM A 325 and
A 490) until these problems with the metric high-strength
bolts are resolved (i.e., standard "English" high-strength
bolts will be the same, except they will have a metric name).
the Standard Specification for Carbon Steel Bolts
and Studs, 414 MPa {60 ksi} Tensile Strength,
ASTM A 307,
the Standard Specification for Structural Bolts,
Steel, Heat-Treated, 827/724 MPa {120/105 ksi}
Minimum Tensile Strength with a required
minimum tensile strength of 827 MPa {120 ksi}
for diameters 12.7 mm through 25.4 mm {1/2 in.
through 1 in.} and 724 MPa {105 ksi} for
diameters 28.6 mm through 38.1 mm {1 1/8 in.
through 1 1/2 in.}, AASHTO M 164 (ASTM
A 325), or
the Standard Specification for Heat-Treated Steel
Structural Bolts, 1034 MPa {150 ksi} Minimum
Tensile Strength, AASHTO M 253 (ASTM
A 490).
AASHTO M 253 (ASTM A 490) bolts are not allowed
unless approved by the Chief Bridge Engineer.
Type 1 bolts should be used with steels other than
weathering steel. Type 3 bolts conforming with either
ASTM A 325 or ASTM A 490 shall be used with
weathering steels. AASHTO M 164 (ASTM A 325), Type
1, bolts may be mechanically galvanized in accordance with
AASHTO M 298 (ASTM B 695), Class 50, when approved
by the Engineer. Galvanized bolts shall be tension tested
A Lehigh University study shows that AASHTO M 253
(ASTM A 490) bolts are more sensitive to the number of
threads in the grip than AASHTO M 164 (ASTM A 325)
bolts. The decrease in tension in AASHTO M 253 (ASTM
A 490) bolts after the maximum tension is reached is much
more rapid than the unloading experienced in the AASHTO
M 164 (ASTM A 325) bolt assembly. Also, the AASHTO
M 253 (ASTM A 490) bolts have reduced ductility
compared to the AASHTO M 164 (ASTM A 325) bolt
B.6 - 4
September 2007
SPECIFICATIONS
COMMENTARY
after galvanizing, as required by AASHTO M 164 (ASTM
A 325).
AASHTO M 253 (ASTM A 490) bolts shall not be
galvanized.
Washers, nuts and bolts of any assembly shall be galvanized
by the same process. The nuts should be overtapped to the
minimum amount required for the fastener assembly, and
shall be lubricated with a lubricant containing a visible dye.
having the same length of thread in the grip. Hot-dipped
galvanized bolts are not permitted due to concerns
associated with the quality of the threads.
6.4.3.2 NUTS
C6.4.3.2
Except as noted below, nuts for AASHTO M 164 (ASTM
A 325) bolts shall conform to either the Standard
Specification for Carbon and Alloy Steel Nuts, AASHTO
M 291 (ASTM A 563), Grades DH, DH3, C, C3, and D, or
the Standard Specification for Carbon and Alloy Steel Nuts
for Bolts for High-Pressure and High-Temperature Service,
AASHTO M 292 (ASTM A 194), Grades 2 and 2H.
Nuts for AASHTO M 253 (ASTM A 490) bolts shall
conform to the requirements of AASHTO M 291 (ASTM
A 563), Grades DH and DH3 or AASHTO M 292 (ASTM
A 194), Grade 2H.
Nuts to be galvanized shall be heat treated, Grade DH .
The provisions of D6.4.3.1 shall apply.
Plain nuts shall have a minimum hardness of 89 HRB.
Nuts to be used with AASHTO M 164 (ASTM A 325),
Type 3 bolts shall be of Grade C3 or DH3. Nuts to be used
with AASHTO M 253 (ASTM A 490), Type 3, bolts shall
be of Grade DH3.
Following the same logic as given in DC6.4.3.1, a soft
metric conversion of standard "English" nuts (ASTM A 563
and ASTM A 194) will be used.
6.4.3.3 WASHERS
Washers shall conform to the Standard Specification for
Hardened Steel Washers, AASHTO M 293 (ASTM F 436).
The provisions of D6.4.3.1 shall apply to galvanized
washers.
6.4.3.4 ALTERNATIVE FASTENERS
The following shall replace the first portion of the first
sentence of A6.4.3.4.
Other fasteners or fastener assemblies, not specified
heretofore, may not be used unless approved by the Chief
Bridge Engineer,
6.4.3.5 LOAD INDICATOR DEVICES
For additional requirements concerning load indicator
devices, see Publication 408.
B.6 - 5
September 2007
SPECIFICATIONS
COMMENTARY
6.4.7 Stainless Steel
The following shall be added to the last sentence of the last
paragraph of A6.4.7.
A...and approved by the Chief Bridge Engineer.@
6.5 LIMIT STATES
6.5.2 Service Limit State
C6.5.2
The following shall replace the second paragraph of A6.5.2
Flexural members shall be investigated at the service limit
state as specified in A6.10, D6.10, A6.11, D6.11, DE6.10P
and DE6.11P.
The following shall replace AC6.5.2
The intent of the service limit state provisions specified for
flexural members in A6.10, D6.10, A6.11, D6.11, DE6.10P
and DE6.11P is primarily to prevent objectionable
permanent deformations due to localized yielding that
would impair rideability under expected severe traffic
loadings.
6.5.3 Fatigue and Fracture Limit State
The following shall replace the third paragraph of A6.5.3
Flexural members shall be investigated at the fatigue and
fracture limit state as specified in A6.10, D6.10, A6.11,
D6.11, DE6.10P and DE6.11P.
6.5.4 Strength Limit State
6.5.4.2 RESISTANCE FACTORS
C6.5.4.2
Replace all references to A 325M and A 490M with A 325
and A 490, respectively.
The following shall replace the pile resistance factors in
A6.5.4.2.
The basis for the resistance factors for driven steel piles is
described in DC6.15.2.
for axial resistance of piles in compression and subject
to damage due to severe driving conditions where use
of a pile tip is necessary .................................. φc = 0.35
for axial resistance of piles in compression under good
driving conditions where use of a pile tip is not
necessary......................................................... φc = 0.45
for axial resistance of piles bearing on soluble bedrock
φc = 0.25
for axial resistance of steel portion of concrete filled
pipe piles in compression................................ φc = 0.35
for combined axial
undamaged piles:
and
flexural
resistance
of
axial resistance .................................φc = 0.60
B.6 - 6
September 2007
SPECIFICATIONS
COMMENTARY
flexural resistance ............................. φf = 0.85
6.6 FATIGUE AND FRACTURE CONSIDERATIONS
6.6.1 Fatigue
6.6.1.2 LOAD-INDUCED FATIGUE
6.6.1.2.1 Application
C6.6.1.2.1
Delete the second sentence of the first paragraph of
A6.6.1.2.1
Delete the first paragraph of AC6.6.1.2.1.
6.6.1.2.2 Design Criteria
For load-induced fatigue considerations, each detail shall
satisfy:
PTF
f
F
n
(6.6.1.2.2-1)
where:
γ
=
load factor specified in Table A3.4.1-1 for the
fatigue load combination
(Δf) =
the force effect, live load stress range due to the
passage of the fatigue load as specified in A3.6.1.4
(MPa) {ksi}
(ΔF)n= the nominal fatigue resistance as specified in
A6.6.1.2.5 and D3.6.1.4 (MPa) {ksi}
=
PTF =
½(ΔF)TH as specified in A6.6.1.2.5 and D3.6.1.4 for
Interstate and NHS bridges
Pennsylvania Traffic Factor as given in Table 1
B.6 - 7
September 2007
SPECIFICATIONS
COMMENTARY
Table 6.6.1.2.2-1 - Pennsylvania Traffic Factor for
Cases I, II and III
Case
Type of Road
ADTT
PTF
---
1.2
I
National Highway
System
II
Freeways,
Expressways,
Major Highways
and Streets
> 500
1.2
III
Other Highways
and Streets not
included in Case I
or II
---
1.0
6.6.1.2.4 Restricted Use Details
C6.6.1.2.4P
Details defined as Category D or E in Table A6.6.1.2.3-1
are considered unacceptable for new designs. Such details
shall be excluded from new designs, except when approved
by the Chief Bridge Engineer.
Girder or floorbeam flanges inserted through a slot cut in
the web of an intersecting member and then welded to one
or both sides of the web to provide continuity are not
acceptable. Moreover, such flanges butted flush against the
web of the intersecting member and then welded to it are
unacceptable.
Details involving the intersection of the flange of one girder
with the web of another girder are unacceptable because a
significant embedded crack-like interface may remain
between members after the welding. Such a defect can
quickly propagate, causing premature failure.
6.6.1.2.5 Fatigue Resistance
C6.6.1.2.5
Replace all references to A 325M and A 490M with A 325
The following shall replace Equation A6.6.1.2.5-2.
The following shall replace fourth paragraph of
AC6.6.1.2.5.
PennDOT's design life is considered to be 100 years. In the
overall development of the LRFD Specification, the design
life has been considered to be 75 years. This is the reason
that the 75 in Equation A6.6.1.2.5-2 is replaced with 100 in
Equation D6.6.1.2.5-2P.
N = (365) (100) n (ADTT)SL
(6.6.1.2.5-2P)
6.6.1.3 DISTORTION-INDUCED FATIGUE
C6.6.1.3
A6.6.1.3.
To control web buckling and elastic flexing of the web, the
provision of A6.10.5.3, D6.10.5.3 and DE6.10.6P shall be
satisfied.
The interaction of primary and secondary components of
steel bridge structures often results in cracking at
unexpected locations in relatively short periods of time.
Such cracking was first observed in the webs of girder-type
bridges at short gaps between transverse web attachments
B.6 - 8
September 2007
SPECIFICATIONS
COMMENTARY
and the girder flanges. Investigations of this type of crack
development suggest that the cracking is typical and is
caused by out-of-plane displacements which result in large
secondary web bending stresses. This is evident in the indepth case studies presented by Mertz (1984).
Fatigue crack growth resulting from displacement-induced
secondary stresses is difficult to anticipate, since it involves
the actual behavior of a structure, rather than the assumed
behavior. The differences between the actual and the
assumed behavior are most critical at very localized regions,
such as at the ends of cut-short transverse connection plates.
The present design idealization does not account for such
localized behavior.
6.6.1.3.4P Distortion-Induced Fatigue: Unacceptable
Details and Acceptable Alternative Details
C6.6.1.3.4P
Members and fasteners shall be detailed to reduce the effect
of repeated variations or reversals of stress due to out-ofplane deformations or secondary forces. Examples of
details which have proven to be unacceptable, based upon
this criteria, are shown in Figure 1. Acceptable alternatives
to these unacceptable details are shown in Figure 2. These
details do not include all possible variations of distortionsensitive details, but they are considered typical and will
provide guidance.
Rather than attempting to quantify the displacement-induced
stresses and develop allowable values, it is the Department's
philosophy that details susceptible to out-of-plane distortion
are not acceptable. Through the design of better details, the
inadequacy of the present design idealization in dealing with
displacement-induced stresses is minimized.
Connection plates for either diaphragms or floorbeams shall
be rigidly attached to both girder flanges (either bolted or
welded). Cutting the connection plate short or merely
providing a tight fit to the flange is not acceptable, since the
potential for localized out-of-plane distortion cracking of the
web exists near the juncture of the web and flange.
Lateral gusset plates near transverse stiffeners or coped
around transverse stiffeners shall be rigidly attached to the
transverse stiffener (either bolted or welded). If this rigid
attachment is not provided, the potential for localized outof-plane distortion cracking of the web is created near the
juncture of the web and transverse stiffener.
If lateral bracing is required, the preferred approach is to
attach the gusset plate to the flange as shown in BD-620M
and BC-754M. Welding of the gusset plate to the stiffener
must be detailed to prevent intersecting walls.
B.6 - 9
September 2007
Figure 6.6.1.3.4P-1 - Unacceptable Details
B.6 - 10
September 2007
Figure 6.6.1.3.4P-1 - Unacceptable Details (Continued)
B.6 - 11
September 2007
Figure 6.6.1.3.4P-1 - Unacceptable Details (Continued)
B.6 - 12
September 2007
Figure 6.6.1.3.4P-2 - Acceptable Alternatives Details
B.6 - 13
September 2007
Figure 6.6.1.3.4P-2 - Acceptable Alternatives Details (Continued)
B.6 - 14
September 2007
Figure 6.6.1.3.4P-2 - Acceptable Alternatives Details (Continued)
B.6 - 15
September 2007
SPECIFICATIONS
COMMENTARY
6.6.2 Fracture
Charpy V-Notch tests shall be performed as specified as per
Publication 408, Section 1105.02(a)4. Diaphragms, crossframes, bracing and connecting plates for curved girder
bridges, straight girder bridges with skew less than 70 , or
connections which are entirely welded and without any
bolting are to be Charpy V-Notch tested. Typical shop
welded, field bolted diaphragms on straight bridges do not
require Charpy V-Notch testing (unless bridge skew is less
than 70 ). Under full dead load, beam ends and all bearing
stiffeners, including bearing stiffeners at piers, are to be
vertical.
6.7 GENERAL DIMENSION AND DETAIL
REQUIREMENTS
C6.7.2P
6.7.2 Dead Load Camber
Camber is provided for the beams so that after all the dead
loads (not including the future wearing surface) are applied,
the beam is at the proper elevation. Camber is not used for
the control of live load deflections.
For curved girders, the designer shall add the following note
in the General Note section of the plans:
“Girder webs shall be plumb under the full dead load
existing at the end of construction”
Full dead load includes the weight of pavement or overlays
included in the initial construction. It does not include the
future wearing surface.
6.7.2.1P CAMBER DUE TO WEIGHT OF DECK SLAB
The camber, due to the weight of deck slab, shall be
determined from an analysis in which the weight of the deck
slab is applied all at once.
6.7.2.2P CAMBER DETAILS FOR DESIGN DRAWINGS
A diagram and a table of camber ordinates (see Figure 1)
shall be shown on the contract plans. Ordinates shall be
provided for all beams at one-tenth points and at field splice
points (at dead load points of contraflexure if field splices
are not provided) to account for the following:
(a) Weight of steel
(b) Weight of deck slab (see D6.7.2.1P)
(c) Superimposed dead load (do not include future
wearing surface)
(d) Vertical curve
(e) Superelevation
B.6 - 16
SPECIFICATIONS
September 2007
COMMENTARY
(f) 50% of heat curve camber (see D6.7.2.3P)
(g) Total due to above
Figure 6.7.2.2P-1 - Camber Details
B.6 - 17
September 2007
SPECIFICATIONS
COMMENTARY
When total camber is less than the minimum that can be
maintained in a beam (WF sections), no camber is required,
but the following note shall be shown on the contract plans:
“Beams shall be placed with any mill camber up;
the contractor shall consider and compensate for
dead load deflection, due to the weight of the
concrete, when forming and constructing the deck
slab.”
Designers shall show theoretical dead load deflection data
on plans even when no special camber is to be fabricated
into the beams (i.e., when using mill camber), since this
information is required by the contractor to construct the
deck to the correct finished deck elevation.
The requirements for cross-section elevations at 3000 mm
{10 ft.} intervals along the length of girder bridges are
found in PP1.6.4.10.
6.7.2.3P HEAT-CURVE CAMBER CORRECTIONS
C6.7.2.3P
To compensate for possible loss of camber of heat-curved
girders in service as residual stresses dissipate, the amount
of camber in mm {in.}, Δ, at any section along the length of
the girder shall be equal to:
Part of the heat-curve camber loss is attributable to
construction loads and will occur during construction of the
bridge; total heat-curve camber loss will be complete after
several months of in-service loads. Therefore, a portion of
the heat-curve camber increase (approximately 50%) should
be included in the bridge profile. Camber losses of this
nature (but generally smaller in magnitude) are also known
to occur in straight beams and girders.
DL
=
(
M
+
R
)
(6.7.2.3P-1)
M
where:
For R < 305 000 mm {1,000 ft.}
Metric Units:
R
=
0.02 L 2 F y
EYo
305 000 - R
260 000
(6.7.2.3P-2)
R
=
0.02 L 2 F y
EYo
1,000 - R
850
For R > 305 000 mm {1,000 ft.}
ΔR =
0
R
=
radius of curvature of curved girder (mm) {ft.}
E
=
modulus of elasticity of the girder flange (MPa)
{ksi}
B.6 - 18
September 2007
SPECIFICATIONS
COMMENTARY
Fy =
specified minimum yield strength of the girder
flange (MPa) {ksi}
Yo =
distance from the neutral axis to the extreme outer
fiber (mm) {in.}
span length for simple span; the distance between a
simple end support and the dead load contraflexure
point or the distance between points of dead load
contraflexure points for continuous spans (mm)
{in.}
L
=
ΔDL =
camber at any point along L (mm) {in.}
ΔM =
maximum value of ΔDL within L (mm) {in.}
Heat-curve camber loss between dead load contraflexure
points adjacent to piers is small and may be neglected.
6.7.3 Minimum Thickness of Steel
C6.7.3P
The following shall replace the first two paragraphs of
A6.7.3.
Structural steel (including bracing, cross-frames, and all
types of gusset plates), except for webs of certain rolled
shapes, closed ribs in orthotropic decks, fillers and in
railings shall not be less than 10 mm {3/8 in.} in thickness.
For girders, the minimum flange plate thickness shall be 20
mm {3/4 in.} unless the fabricator can demonstrate the
ability to satisfactorily fabricate and erect plate girders with
thinner flange plates. For girders with longitudinal
stiffeners, the minimum web thickness shall be 12 mm {1/2
in.}. The web thickness of rolled beams or channels shall
not be less than 6 mm {0.23 in.}. The thickness of closed
ribs in orthotropic decks shall not be less than 5 mm {3/16
in.}.
For girder flanges, bearing stiffeners and splice plates for
bridges that are to be metallized, the width of the plates are
to be oversized by 3 mm {1/8 in.} to account for edge
grinding. The flange, bearing stiffener plates and splice
plates to be shown on the plans shall be the oversized plates.
For metallized bridges, the estimated quantity of fabricated
structural steel shall be based on the oversized plates.
This requirement of minimum web thickness for girders
with longitudinal stiffeners was added to avoid web
buckling and oil canning of deep girders. PennDOT has
previously used 10 mm {3/8 in.} thickness resulting in web
oil canning effect, specifically on I-476 over Conestoga
Avenue. The New York Department of Transportation has
successfully used the specified criteria.
For metallized bridges, the rolled edges of angles, channels
and wide flange beams do not require edge grinding,
therefore these components are not to be oversized.
6.7.4 Diaphragms and Cross-Frames
6.7.4.1 GENERAL
C6.7.4.1
A6.7.4.1.
The maximum spacing of cross-frames or diaphragms shall
be 7600 mm {25 ft.}.
Skew effects must be considered when designing
diaphragms, especially when the skew angle is less than
B.6 - 19
September 2007
SPECIFICATIONS
COMMENTARY
70 . Proper consideration of unbraced length and
diaphragm loads from non-uniform deflections is
mandatory. Design calculations must consider the fact that
cross-frames in skewed bridges connect different points of
the span of adjacent girders and that these points will not
deflect the same amount. Therefore, a check considering
these differences must be made, and the resulting design
forces must be used in the cross-frame design.
For sharply skewed bridges (typically, skews less than or
equal to 60 ), a cross-frame or diaphragm normal to the
girder shall be located such as to minimize the effects of
differential deflections, while satisfying the minimum crossframe or diaphragm spacing requirement.
For additional analysis criteria for bridges with skew angles
less than 70 , see D4.6.2.2.1.
Diaphragm and cross-frame members in horizontally curved
bridges shall be considered primary members.
bridges may be used at support locations.
In locating intermediate cross-frames or diaphragms in
sharply skewed brides, the designer must consider distinct
issues associated with each girder connected by the crossframe or diaphragm. A cross-frame or diaphragm close to
the bearing on one girder line may introduce forces into the
system (cross-frame or diaphragm and girder flange) due to
―nuisance stiffness,‖ where the deflection of one girder line
cannot match the adjacent girder line. In these cases,
elimination of a cross-frame or diaphragm is advisable. In
addition, the initial cross-frame or diaphragm must be
located such that the maximum permitted spacing is not
exceeded in the adjacent connected girder. In some cases,
the first interior line of cross-frames or diaphragms may not
be full width across the superstructure, and the number of
bays along a girder length may not be constant for each
girder in the superstructure.
Bracing of horizontally curved members is more critical
than for straight members. Diaphragm and cross-frame
members resist forces that are critical to the proper
functioning of curved-girder bridges. Since they transmit the
forces necessary to provide equilibrium, they are considered
primary members. Therefore, forces in the bracing members
must be computed and considered in the design of these
members. When the girders have been analyzed neglecting
the effects of curvature according to the provisions of
A4.6.1.2.4, the diaphragms or cross-frames may be analyzed
by the V-load method (United States Steel 1984) or other
rational means.
Solid plate diaphragms can be used at support
locations in horizontally curved and skewed bridges.
Replacing cross frames with solid plate diaphragms at
abutment and pier locations has been shown to not
adversely affect or appreciably benefit deformations during
construction. For other intermediate locations along the
bridge spans, diaphragms may cause higher stresses and
deformations in the bridge structures during construction
when compared to the use of cross frames.
When support lines are skewed less than 70 degrees,
intermediate diaphragms or cross-frames shall be placed
normal to the girders in contiguous or discontinuous lines.
When cross frames and diaphragms are normal to web near
skewed supports, adequate girder restraint shall be
provided.
Placement of cross frames parallel to the skew has been
shown to induce significant localized lateral bending near
support locations (AASHTO Subcmte on Bridges &
Structures, 2010).
6.7.4.2 I-SECTION MEMBERS
C6.7.4.2P
B.6 - 20
September 2007
SPECIFICATIONS
COMMENTARY
The article heading is changed from: Straight I-Sections.
End cross-frames must be parallel to centerline of bearings,
but need not coincide with bearing line.
A6.7.4.2
Cross frames and steel girders in curved and skewed
bridges should both be detailed so that the webs are plumb
under a specified loading condition.
Cross-frame members in horizontally curved bridges should
contain diagonals and top and bottom chords.
The following shall replace the last two sentences of second
When the supports are skewed less than 70°, intermediate
cross-frames shall be normal to the main members. If the
supports are skewed, end cross-frames need not be co-linear
with the line of bearings, see Standard Drawing BC-754M.
For additional skewed cross-frame requirements, see
D6.7.4.1.
In curved and skewed bridges, cross-frame stresses can
increase appreciably as a result of the locked-in stresses
caused by inconsistent detailing. This increase may be more
pronounced in bridges with larger cross frame spacings and
also for cross frames near the bridge supports. For skewed
bridges with skew angles between 90 and 70 , develop
shop drawings which detail all webs plumb when girders
are erected and diaphragms connected. For curved bridges
and skewed bridges with skew angles less than 70 , develop
shop drawings and erection procedures which detail all
webs plumb after the full dead load (self weight of all
structural and non-structural components, not including
weight of the future wearing surface) is applied. See
PennDOT Bridge Construction Standard BC-754, “Steel
Diaphragms.”
The following shall supplement A6.7.4.2
The spacing, Lb, of intermediate diaphragms or crossframes in horizontally curved I-girder bridges shall not
exceed the following in the erected condition:
Lb
Lr
R / 10
(6.7.4.2-1)
where:
Lr =
limiting unbraced length
Eq. A6.10.8.2.3-5 (mm) {ft.}
R
minimum girder radius within the panel (mm) {ft.}
=
determined
from
In no case shall Lb exceed 7500 mm {25 ft}.
6.7.4.3 BOX SECTION MEMBERS
C6.7.4.3
The article heading is changed from: Straight Box Sections.
The following shall replace the first sentence of the fourth
paragraph of A6.7.4.3
For all horizontally curved box girder bridges, single box
sections, and for box sections in bridges not satisfying the
requirements of A6.11.2.3 or with box flanges that are not
fully effective, cross-sectional distortion stresses are best
controlled by the introduction of internal cross-frames or
diaphragms.
The following shall replace the last two paragraphs of
A6.7.4.3.
Intermediate internal diaphragms or cross-frames shall be
provided. For all single box sections, horizontally curved
sections, and multiple box sections in cross-sections of
bridges not satisfying the requirements of A6.11.2.3 or with
box flanges that are not fully effective according to the
B.6 - 21
September 2007
SPECIFICATIONS
COMMENTARY
provisions of A6.11.1.1 and D6.11.1.1 for curved bridges or
DE6.11.1.1P for straight bridges, the internal bracing shall
be spaced to control cross-section distortion, with the
spacing not to exceed 7500 mm {25 ft}.
For all single box sections, horizontally curved sections, and
multiple box sections in bridges not satisfying the
requirements of A6.11.2.3 or with box flanges that are not
fully effective according to the provisions of A6.11.1.1 and
D6.11.1.1 for curved bridges or DE6.11.1.1P for straight
bridges, the need for a bottom transverse member within the
internal bracing shall be considered. Where provided, the
transverse member shall be attached to the box flange unless
longitudinal flange stiffeners are used, in which case the
transverse member shall be attached to the longitudinal
stiffeners by bolting. The cross-sectional area and stiffness
of the top and bottom internal bracing members shall also
not be less than the area and stiffness of the diagonal
members.
B.6 - 22
September 2007
SPECIFICATIONS
COMMENTARY
6.7.5 Lateral Bracing
6.7.5.2 I-SECTION MEMBERS
C6.7.5.2
The article heading is changed from: Straight I-Sections.
Wherever possible, horizontal lateral bracing in or near the
plane of the bottom flange should be eliminated from
bridges, and the girders should be designed to carry the
wind load between diaphragms according to A4.6.2.7.1.
Horizontal lateral bracing is relatively expensive because of
the detail associated with it. Furthermore, there are often
forces associated with horizontal lateral bracing which can
result in distortion-induced fatigue; these forces are also a
significant factor on steel bridges. Therefore, horizontal
lateral bracing should not be considered for the
improvement of redundancy.
When horizontal lateral bracing is required, it should be
attached to the bottom flange wherever practical. (BD620M permits attachment to the top flange.)
For horizontally curved bridges, when the curvature is
sharp and temporary supports are not practical, it may be
desirable to consider providing both top and bottom lateral
bracing to ensure pseudo-box action while the bridge is
under construction. Top and bottom lateral bracing provides
stability to a pair of I-girders.
6.7.5.3 TUB SECTION MEMBERS
C6.7.5.3
The following shall replace the last sentence of the third
For both straight and horizontally curved tub sections, a
full-length lateral bracing system forms a pseudo-box to
help limit distortions brought about by temperature changes
occurring prior to concrete deck placement, and to resist the
torsion and twist caused by any eccentric loads acting on the
steel section during construction. AASHTO (1993) specified
that diagonal members of the top lateral bracing for tub
sections satisfy the following criterion:
The article title is changed from: Straight Tub Sections.
A6.7.5.3.
For horizontally curved girders, a full-length lateral bracing
system shall be provided.
Ad
0.76w
(C6.7.5.3-1)
where:
Ad =
minimum required cross-sectional area of one
diagonal (mm2) {in.2)
w
center-to-center distance between the top flanges
(mm) {in.}
=
Satisfaction of this criterion was intended to ensure that the
top lateral bracing would be sized so that the tub would act
as a pseudo-box section with minimal warping torsional
displacement and normal stresses due to warping torsion
less than or equal to 10 percent of the major-axis bending
stresses. This criterion was developed assuming tub sections
with vertical webs and ratios of section width-to-depth
B.6 - 23
between 0.5 and 2.0, and an X-type top lateral bracing
system with the diagonals placed at an angle of 45° relative
to the longitudinal centerline of the tub-girder flanges
September 2007
SPECIFICATIONS
COMMENTARY
CL
.
Abut
C
L
Abu
t.
Figure C6.7.5.3-1 Warren-Type Single-Diagonal Top
Lateral Bracing System for Tub Section Member: Plan
View.
CL
.
Abut
C
L
Abu
t.
Figure C6.7.5.3-2 Pratt-Type Single-Diagonal Top Lateral
Bracing System for Tub Section Member: Plan View.
Where the forces in the bracing members are not available
from a refined analysis, the shear flow across the top of the
pseudo-box section can be computed from Eq. C6.11.1.1-1
assuming the top lateral bracing acts as an equivalent plate.
The resulting shear can then be computed by multiplying the
resulting shear flow by the width w, and the shear can then
be resolved into the diagonal bracing member(s). Should it
become necessary for any reason to compute the St. Venant
torsional stiffness of the pseudo-box section according to
Eq. AC6.7.4.3-1, formulas are available (Kollbrunner and
Basler 1966; Dabrowski 1968) to calculate the thickness of
the equivalent plate for different possible configurations of
top lateral bracing.
6.8 TENSION MEMBERS
6.8.2 Tensile Resistance
6.8.2.2 REDUCTION FACTOR, U
The following shall replace the first paragraph of A6.8.2.2.
The reduction factors, specified in A6.8.2.2, shall be used to
account for shear lag. Reduction factors developed from
refined analysis or tests may be used if approved by the
6.8.2.3 COMBINED TENSION AND FLEXURE
The following shall replace the definition of Mrx in
A6.8.2.3.
Mrx =
factored flexural resistance about the x-axis taken
as f times the nominal flexural resistance about
the x-axis determined as specified in DE6.10P,
DE6.11P or A6.12, as applicable (N-mm) {kip-in.}
B.6 - 24
September 2007
SPECIFICATIONS
COMMENTARY
6.8.3 Net Area
C6.8.3
The following shall replace the first paragraph of A6.8.3.
The net area, An, of an element is the product of the
thickness of the element and its smallest net width. The net
area, An, of a member is the sum of the net areas of each
element. The width deducted for all holes; standard,
oversize,
and slotted; shall be taken as 1.6 mm {1/16 in.} greater than
the hole size specified in D6.13.2.4.2. The net width shall
be determined for each chain of holes extending across the
member or element along any transverse, diagonal, or
zigzag line.
Delete the first paragraph of AC6.8.3.
6.9 COMPRESSION MEMBERS
6.9.2 Compressive Resistance
6.9.2.2 COMBINED AXIAL COMPRESSION AND
FLEXURE
The following shall replace the definition of Mrx in
A6.8.2.3.
Mrx =
factored flexural resistance about the x-axis taken
equal to f times the nominal flexural resistance
about the x-axis determined as specified in
DE6.10P, DE6.11P or A6.12, as applicable (Nmm) {kip-in.}
6.9.5 Composite Members
6.9.5.1 Nominal Compressive Resistance
The following shall replace the definition of n in
A6.9.5.1.
n
= modular ratio of the concrete as specified in
D5.4.2.1
6.10 I-SECTIONS IN FLEXURE
C6.10
6.10.0P Applicable Provisions
C6.10.0P
The provisions of D6.10 apply to steel I-girders curved in
plan. The provisions of Appendix E apply to straight girder
bridges.
The provisions of D6.10 are based on the 2004 Third
Edition of AASHTO-LRFD specifications as amended
herein to account for the curved girder provisions that were
incorporated in AASHTO LRFD by the 2005 interim
specifications. The provisions of Appendix E are based on
the 1998 Second Edition of the AASHTO-LRFD Design
specifications with the 1999 through 2003 Interims. Using
the earlier specifications for straight girder bridges will
Where Straight girders are referred to in A6.10 and D6.10,
this reference should be ignored.
B.6 - 25
September 2007
SPECIFICATIONS
COMMENTARY
allow continuing using the existing STLRFD computer
program, which is based on the Second Edition of AASHTO
LRFD with modifications, with relatively few revisions. It
is the intent of the Department to update the STLRFD
computer program to correspond to the 2004 AASHTOLRFD, including A6.10 and A6.11, at some point in the
future. At which time, both straight and curved girders will
be designed using the provisions of the 2004 Third Edition
of the Specifications as amended by Design Manual Part 4.
Any reference to straight girder bridges in D6.10 is meant to
be used in the future when the Design Manual Part 4 and
computer programs are fully updated to the third edition of
the AASHTO-LRFD specifications.
6.10.1 General
C6.10.1
The following shall replace the first sentence in A6.10.1
The provisions of this article apply to flexure of rolled or
fabricated kinked (chorded) continuous or horizontally
curved steel I-section members symmetrical about the
vertical axis in the plane of the web.
The following shall replace the first paragraph of AC6.10.1.
This article addresses general topics that apply to all types
of steel I-sections in horizontally curved bridges, or bridges
containing both straight and curved segments. For the
application of the provisions of A6.10 and D6.10, bridges
containing both straight and curved segments are to be
treated as horizontally curved bridges since the effects of
curvature on the support reactions and girder deflections, as
well as the effects of flange lateral bending, usually extend
beyond the curved segments. Note that kinked (chorded)
girders exhibit the same actions as curved girders, except
that the effect of the noncollinearity of the flanges is
concentrated at the kinks. Continuous kinked (chorded)
girders should be treated as horizontally curved girders with
respect to these Specifications.
Open-framed systems are those which have no horizontal
lateral bracing in or near the plane of the bottom flange.
Lateral bracing, when used, is provided to resist wind loads,
but it is generally not needed since the girders can be
For horizontally curved bridges, in addition to the potential
sources of flange lateral bending discussed in the preceding
paragraph, flange lateral bending effects due to curvature
must always be considered at all limit states and also during
construction.
Delete the reference to Appendix B from the first sentence
of the second Paragraph of AC6.10.1.
The following shall replace the second sentence of
the second paragraph of A6.10.1.
For the majority of straight non-skewed bridges, flange
lateral bending effects tend to be most significant during
construction and tend to be insignificant in the final
constructed condition. Significant flange lateral bending
may be caused by wind, by torsion from eccentric concrete
deck overhang loads acting on cantilever forming brackets
placed along exterior girders, and by the use of staggered
cross-frames in conjunction with skews less than 70°.
The application of open-framed system distribution factors
for closed-framed systems is generally conservative.
B.6 - 26
September 2007
SPECIFICATIONS
COMMENTARY
designed to carry wind loads between the diaphragms.
If horizontal lateral bracing is included, the open-framed
system distribution factors shall be used. If a horizontal
lateral bracing system is used, the connections must be
detailed to ensure that the fatigue life of the bracing system
is at least that of the girder.
Although the lateral wind bracing may not be required for
the final constructed condition, the need for lateral wind
bracing during construction shall be investigated.
If horizontal lateral bracing system is used, a rational
analysis may consider a reduction in lateral live load
distribution factor due to the quasi-box action of the closedframe system.
Any reduction in live load distribution factor must be
approved by the Chief Bridge Engineer.
The design procedure for evaluating the need for lateral
bracing during construction shall be per BD-620M. As
agreed upon by the APC Subcommittee for Steel Bridge
Superstructures, the contractor is responsible for stability of
the girders during erection, including providing wind
bracing during erection as needed. This responsibility
includes the analysis, design, material, fabrication and
installation (and removal) of wind bracing during erection at
no cost to the Department.
6.10.1.1.1 Stresses
C6.10.1.1.1P
The following shall supplement A6.10.1.1.1
If concrete with expansive characteristics (except shrinkagecompensating concrete, which the Department is studying
and may eventually exclude from this provision) is used,
composite design shall be used with caution, and provision
must be made in the design to accommodate the expansion.
Composite section properties (see D6.10.3.1.1b) shall be
assumed in the positive and negative moment regions for
the calculation of design moments, shears and deflections.
If the concrete is expansive, estimate expansion and
properly design concrete to flange connection by adding
additional shear studs.
6.10.1.1.1a Sequence of Loading
C6.10.1.1.1a
The following shall replace last paragraph of A6.10.1.1.1a
For unshored construction, permanent load applied before
the concrete deck has attained 75% of its compressive
strength shall be assumed carried by the steel section alone;
permanent load and live load applied after this stage shall be
assumed carried by the composite section. For shored
construction, all permanent loads shall be assumed applied
after the concrete deck has hardened or has been made
composite and the contract documents shall so indicate.
Use of shored systems requires the prior approval of the
For continuous spans, the final dead load moment at each
design section shall be taken as the greater of either the dead
load moment considering the weight of the concrete deck to
be instantaneously applied or a moment based upon an
incremental analysis of the specified slab placement
sequence. Similarly, stresses should be computed based on
the more critical of the incremental and instantaneously
applied loads.
Delete the first paragraph of AC6.10.1.1.1
6.10.1.1.1b Stresses for Sections in Positive Flexure
C6.10.1.1.1b
Delete Equation A6.10.1.1.1b-1
The following shall replace AC6.10.1.1.1b
B.6 - 27
September 2007
SPECIFICATIONS
COMMENTARY
The following shall supplement A6.10.1.1.1b:
For normal and low density concrete, the modular ratio is
given in D5.4.2.1.
It is preferable to proportion composite sections in simple
spans and the positive moment regions of continuous spans
so that the neutral axis lies below the top surface of the steel
beam.
6.10.1.1.1c Stresses for Sections in Negative Flexure
The following shall replace A6.10.1.1.1c
For calculating flexural stresses in sections subjected to
negative flexure, the composite section for both short-term
and long-term moments shall consist of the steel section and
the longitudinal reinforcement within the effective width of
the concrete deck.
Cut-off points for the main reinforcement in cast-in-place
decks over interior supports for continuity may be staggered
as required by design.
Concrete on the tension side of the neutral axis shall not be
considered in calculating resisting moments.
6.10.1.1.1d Concrete Deck Stresses
C6.10.1.1.1d
The following shall replace A6.10.1.1.1d
For calculating longitudinal flexural stresses in the concrete
deck due to transient loads, the short-term modular ratio, n,
shall be used. For calculating longitudinal flexural stresses
in the concrete deck due to permanent loads, the long-term
modular ratio, 3n, shall be used.
Delete AC6.10.1.1.1d
6.10.1.1.1.fP Lateral Support of Top Flanges Supporting
Timber Decks
The compression flanges of girders supporting timber floors
shall not be considered to be laterally supported by the
flooring, unless the floor and fastenings are specially
designed to provide such support. Laminated timber decks
shall be provided with steel clips designed to furnish
adequate lateral support to the top flange.
6.10.1.2 NONCOMPOSITE SECTIONS
Whenever technically feasible, all structures shall be made
composite.
B.6 - 28
September 2007
SPECIFICATIONS
COMMENTARY
6.10.1.3 HYBRID SECTIONS
The use of girders with web yield strength higher than the
flange yield strength requires the prior approval of the Chief
Bridge Engineer.
6.10.1.4 VARIABLE WEB DEPTH MEMBERS
The use of girders with variable web depth requires the prior
6.10.1.5 STIFFNESS
C6.10.1.5
In the computation of flexural stiffness and flexural
resistance of beams, the haunch shall be taken as zero.
However, in the computation of dead load, the haunch shall
be taken into account.
The following shall supplement AC6.10.1.5
Field measured haunch depths may be used in the
computation for flexural stiffness and resistance when rating
existing bridges.
AC6.10.1.5.
Field tests of composite continuous bridges have shown that
there is considerable composite action in negative bending
regions (Baldwin et al. 1978; Roeder and Eltvik 1985; Yen
et al. 1995). Therefore, the stiffness of the full composite
section is to be used over the entire bridge length for the
analysis of composite flexural members, but not for stress
calculations.
Other stiffness approximations which are based on sound
engineering principles may be used if approved by the Chief
Bridge Engineer.
6.10.1.6 FLANGE STRESSES AND MEMBER
BENDING MOMENT
C6.10.1.6
The following shall replace the forth paragraph of A6.10.1.6
to the end of the article.
Lateral bending stresses in continuously braced flanges shall
be taken equal to zero. Lateral bending stresses in discretely
braced flanges shall be determined by structural analysis.
All discretely braced flanges shall satisfy:
The following shall supplement the fourth paragraph of
AC6.10.1.6
The determination of flange lateral bending moments due to
curvature is addressed in A4.6.1.2.4b.
In all resistance equations, fbu, Mu and f are to be taken as
positive in sign. However, for service and strength limit
state checks at locations where the dead and live load
contributions to fbu, Mu or f a re of opposite sign, the signs
of each contribution must be initially taken into account. In
such cases, for both dead and live load, the appropriate net
sum of the major-axis and lateral bending actions due to the
factored loads must be computed, taking the signs into
consideration, that will result in the most critical response
for the limit state under consideration.
f
(6.10.1.6-1)
0.6Fyf
The flange lateral bending stress, f , may be determined
directly from first-order elastic analysis in discretely braced
compression flanges for which:
Lb
1.2 Lp
Cb Rb
fbu / Fyc
(6.10.1.6-2)
B.6 - 29
September 2007
SPECIFICATIONS
COMMENTARY
or equivalently:
Lb
1.2 Lp
Cb Rb
M u / M yc
(6.10.1.6-3)
where:
Cb =
moment gradient modifier specified in A6.10.8.2.3,
D6.10.8.2.3 or Appendix A Article A6.3.3, as
applicable.
fbu =
largest value of the compressive stress throughout
the unbraced length in the flange under
consideration, calculated without consideration of
flange lateral bending (MPa) {ksi}
Lb =
unbraced length (mm) {in.}
Lp =
limiting unbraced length specified in A6.10.8.2.3
and D6.10.8.2.3 (mm) {in.}
Mu =
largest value of the major-axis bending moment
throughout the unbraced length causing
compression in the flange under consideration (Nmm) {kip-in.}
Myc =
yield moment with respect to the compression
flange determined as specified in AD6.2 (N-mm)
{kip-in.}
Rb =
web load-shedding factor determined as specified
in A6.10.1.10.2
If Eq. 2, or Eq. 3 as applicable, is not satisfied, second-order
elastic compression-flange lateral bending stresses shall be
determined.
Second-order compression-flange lateral bending stresses
may be determined by amplifying first-order values as
follows:
f
0.85
f
1 bu
Fcr
f
f
1
(6.10.1.6-4)
1
or equivalently:
f
0.85
Mu
1
Fcr S xc
f
1
f
1
(6.10.1.6-5)
B.6 - 30
September 2007
SPECIFICATIONS
COMMENTARY
where:
fbu =
largest value of the compressive stress throughout
the unbraced length in the flange under
consideration, calculated without consideration of
flange lateral bending (MPa) {ksi}
f1 =
first-order compression-flange lateral bending
stress at the section under consideration, or the
maximum first-order lateral bending stress in the
compression flange under consideration throughout
the unbraced length, as applicable (MPa) {ksi}
Fcr =
elastic lateral torsional buckling stress for the
flange under consideration determined from Eq.
6.10.8.2.3-8 or Eq. AA6.3.3-8. Eq. AA6.3.3-8 may
only be applied for unbraced lengths in straight Igirder bridges in which the web is compact or
noncompact.
Mu =
largest value of the major-axis bending moment
throughout the unbraced length causing
compression in the flange under consideration (Nmm) {kip-in.}
Sxc =
elastic section modulus about the major axis of the
section to the compression flange taken as Myc/Fyc
(mm3) {in.3}
6.10.1.7 MINIMUM NEGATIVE FLEXURE CONCRETE
DECK REINFORCEMENT
C6.10.1.7
The following shall replace the first paragraph of A6.10.1.7
In negative flexure regions of any continuous span, the total
cross-sectional area of the longitudinal reinforcement shall
not be less than 1 percent of the total cross-sectional area of
the slab. The reinforcement used to satisfy this requirement
shall have a specified minimum yield strength not less than
420 MPa {60 ksi} and a size not exceeding No. 19 bars
{No. 6}.
The required reinforcement shall be placed in two layers
uniformly distributed across the slab width, and two-thirds
shall be placed in the top layer. The individual bars shall be
spaced at intervals not exceeding 300 mm {12 in} within
each row.
Shear connectors shall be provided along the entire length
of the girder to develop stresses on the plane joining the
concrete and steel in accordance with A6.10.10 and
D6.10.10.
Delete the second paragraph and last paragraph of A6.10.1.7
The following shall replace the last sentence in the
paragraph before last in CA6.10.1.7.
The above applies for members that are designed by the
provisions of Article A6.10, Article D6.10 or Appendix A.
B.6 - 31
September 2007
SPECIFICATIONS
COMMENTARY
6.10.1.11P LATERAL SUPPORT OF TOP FLANGES
SUPPORTING TIMBER DECKS
The compression flanges of girders supporting timber floors
shall not be considered to be laterally supported by the
flooring, unless the floor and fastenings are specially
designed to provide such support. Laminated timber decks
shall be provided with steel clips designed to furnish
adequate lateral support to the top flange.
6.10.3.2 FLEXURE
6.10.3.2.1 Discretely Braced Flanges in Compression
C6.10.3.2.1
The following shall replace the definition of Fnc following
Equation A6.10.3.2.1-3.
The following shall supplement the fourth paragraph
of AC6.10.3.2.1.
For horizontally curved bridges, flange lateral bending
effects due to curvature must always be considered in
discretely braced flanges during construction.
Delete the eighth paragraph of AC6.10.3.2.1.
Fnc =
nominal flexural resistance of the flange (MPa).
Fnc shall be determined as specified in A6.10.8.2
and D6.10.8.2.. In computing Fnc for
constructibility, the web load-shedding factor, Rb,
shall be taken as 1.0.
6.10.3.2.5.1P Slab Placement
C6.10.3.2.5.1P
An analysis shall be performed to determine an acceptable
slab placement sequence. The analysis shall address (but is
not limited to) the following items:
During the mid-1980's, several of the Department's girder
bridges experienced problems during placement of the slab.
It is believed that bridges with highly unsymmetrical, deep
steel girders combined with wide beam spacing and large
overhang dimensions are more susceptible to problems
during construction than are the typical earlier steel girder
bridges which use more nearly symmetrical steel girders
combined with closer beam spacing and smaller overhang
dimensions. Since significant reduction in the construction
cost of a bridge can be achieved by use of highly
unsymmetrical, deep steel girders in conjunction with wide
beam spacing and large overhang dimensions, an analysis
must be performed to ensure that these types of girders
provide adequate stability and strength through slab
placement.
(a) Change in the stiffness in the girder as different
segments of the slab are placed and as it affects
both the temporary stresses and the potential for
"locked-in" erection stresses
(b) Bracing (or lack thereof) of the compression flange
of girders and its effect on the stability and strength
of the girder
(c) Stability and strength of the girder through slab
placement
(d) Bracing of overhang deck forms
(e) Uplift at bearings
(f) Temperature changes as prescribed in 3.12.2.1.1.
With skewed, curved, and/or continuous steel girder
bridges, temporary uplift conditions at bearings can occur
during the deck pour. Designers should evaluate the
potential for uplift in bearings as part of the deck pour
sequence evaluation. Designers should address temporary
uplift conditions as follows:
The analysis of slab placement shall be done in an
incremental fashion using a concrete modulus of elasticity
Where the temporary uplift is not detrimental to the
B.6 - 32
September 2007
SPECIFICATIONS
COMMENTARY
equal to 70% of the concrete modulus elasticity at 28 days
for concrete which is at least 24 hours old, assuming no
retarder admixture is permitted. If retarder admixture is
specified, it shall be indicated on the contract drawings, and
the analysis shall be completed assuming 48 hours before
gaining stiffness for lateral resistance. This means the
stiffness of the model will change at the many different
stages.
In no case shall the final design moment stresses or forces
be less than those determined from an analysis in which the
weight of the deck slab is applied all at once.
Slab concrete, which is less than 24 hours old (or 48 hours
old when retarder is used), cannot be considered to provide
lateral support for the embedded top flange of the girder.
Conversely, slab concrete which is more than 24 hours old
(or 48 hours old when retarder is used) can be considered to
provide full lateral support for the embedded top flange of
the girder. If the contractor can demonstrate that the
concrete will provide lateral support for the embedded top
flange in less than 24 hours (or 48 hours old when retarder
is used), that limiting time may be used with the approval of
From the results of the analysis of slab placement and lateral
support conditions described above, the bending and shear
strength of girder shall be checked.
long-term performance of the bearing, or does not result
in adverse stability conditions, temporary uplift is
permitted. In this case, the designer should identify in
the construction plans the individual bearing locations
where uplift is expected and during what stages of the
deck pour the uplift will occur. A note stating that the
uplift is temporary and permitted as part of construction
should also be provided in the construction plans.
Where uplift is determined to be unacceptable for
individual bearing types or structure stability, the
designer should identify in the construction plans the
individual bearing locations where uplift is expected.
Hold down forces and any other design requirements
for restraining devices should be shown in the plans for
the contractor=s use in designing these components.
Forces and design requirements for individual deck
pour stages, as applicable, should be provided. The
designer should verify the viability of at least one type
of restraining device to meet the design requirements
and provide schematic details of the device in the
construction plans.
The effects from temperature change on the curved and
skewed bridges are mostly functions of the girder support
conditions. When minimum required restraint necessary for
girder(s) global stability (i.e., prevention of global buckling
of the girder or collection of girders) is provided, the
applied temperature change may not have an appreciable
impact on overall bridge deflections and stresses.
6.10.3.2.5.2P Deck Slab Overhang Form Support
C6.10.3.2.5.2P
For the erection condition with the overhang form support
system, the strength and stability of the fascia girder shall be
ensured by applying the dead load of the overhang concrete
and any construction equipment to the girder as follows:
The requirements of this article can be met by reducing the
length of some deck pours, or by increasing the size of the
steel girder section, or by a combination of both. For
original designs, the designer should obtain input from
contractor and fabricators about the economics of those
alternatives. Note, also, that only a relatively short length of
the critical spans will be affected by the constructibility
criterion.
The intent of the required checks is to control the buckling
of the flanges and the webs of steel girders. It is felt that
there is a potential for fatigue cracking if steel plates are
allowed to buckle due to "oil-canning" effects.
The preferred upper limit on the deck slab overhang is
1200mm {4'-0"} considering factors such as deck forming
(a) The standard form support system, shown in Figure 1,
may be used where:
(1) Girder web depth is less than 2400 mm {8'-0"}
(2) Deck slab overhang is less than 1400 mm {4'-9"}
(3) Slab thickness is equal to or less than 250 mm {10
in.}
B.6 - 33
September 2007
SPECIFICATIONS
COMMENTARY
and deck finishing.
(4) Transverse stiffener spacing does not exceed the
depth of the girder
(5) In regions where γw (see A6.10.3.2.1, D6.10.3.2.1
and A6.10.3.2.2) is less than 2.5, the factored dead
load shear using a load factor of 4.0 is less than the
buckling shear given in A6.10.9.3.
B.6 - 34
September 2007
SPECIFICATIONS
COMMENTARY
The fascia girders are designed for a temporary
construction load applied to the web at a maximum
1220mm{4 ft.} interval. This load (see table) approximates
the horizontal component of a deck overhang form support
bracket and consists of an allowance for the weight of the
concrete, forms and incidental loads, plus the deck finishing
machine. Where a transverse stiffener spacing, less than
that required for the final design shear, is indicated for
constructibility, the spacing for the final design shear may
be used if the overhang forms are supported from the
bottom flange of the fascia girder, or if the girder web is
adequately braced to prevent buckling due to loads from
web-bearing form support brackets. The contractor has the
option to modify the overhang bracket from that described
herein provided working drawings including calculations,
sealed by a professional engineer licensed in the
Commonwealth of Pennsylvania, are submitted for review
and acceptance and show the modifications do not cause
unacceptable deformations or stresses in the bridge and it is
understood the contractor is ultimately responsible for the
satisfactory completion of the bridge.
Figure 6.10.3.2.5.2P-1 - Typical Overhang
Forming Detail and Note
Where these requirements are satisfied, original designs of
fascia girders shall provide transverse stiffeners throughout
For rolled beams spans, this is typically not a controlling
design consideration with small deck overhangs less than
B.6 - 35
September 2007
SPECIFICATIONS
COMMENTARY
the span at a maximum spacing of D, including the region
where stiffeners are not required for the final design shear or
where a spacing larger than D would be satisfactory for the
final design shear. This requirement ensures reasonable
constructibility. The stiffener spacings required for both
constructibility and final design shear shall be shown on the
contract drawings (preferably on the girder elevations), and
the sketch and note from Figure 1 shall be included on the
contract drawings.
600 mm {2 ft.} overhangs.
The revision to this note was developed by an APC
Subcommittee for Stability of Steel Bridge Superstructures.
The note was modified to provide more flexibility to the
contractor to use deck overhang form brackets that have
nominal depths greater than the typical 915 mm {3’-0‖}
bracket depth. The maximum permissible horizontal load
value was developed based on field measurements of steel
bridges constructed in 1999 in District 5-0 with deck
overhangs in the range of 1422 mm {4=-8@} and a limited
finite element analysis study of lateral web deflections of
steel girders subjected to concentrated horizontal forces on
the girder web.
Design modifications should consider web stress, overall
web deformation, relative web deformation, the resulting
deck overhang deflection, and the resulting effects on the
finished deck profile. The contractor is responsible for
selecting and providing calculations for the overhang
forming system as required by Publication 408 Section
1050.3(c)2. Publication 408 Section 105.01 (c) specifies the
responsibility of the work remains with the contractor
regardless of reviews and/or acceptance of submitted
working drawings by the Department.
Unacceptable deformations of the web or top flange results
in deflection of the overhang bracket causing problematic
deck finish and ride quality.
(b) For deck slab overhangs which do not meet the
requirements of (a), the designer of the original
structure shall review the condition with the Chief
Bridge Engineer's office as part of the TS&L
submission. If it is determined that web-supported
overhang form brackets cannot be permitted, the
following note shall be included in the general
notes:
Support deck slab overhang forms from the bottom
flange of the fascia girder, unless the girder web is
adequately supported to prevent buckling due to
loads from web-bearing form supports.
(c)
Contractor-designed alternates shall meet the
requirements of this article. The stiffener spacing
and a description of the deck overhang form
support system, including the loads, shall be shown
on the conceptual design drawings submitted for
approval.
(d)
All DM-4 and appropriate LRFD provisions in
regard to flange and web buckling must be
checked.
(e)
For additional criteria on exterior girder rotation
due to large cantilever deck slabs, see D9.7.1.5.1P.
If an overhang is braced to within 6 inches of the bottom
flange, it shall be considered braced to the bottom flange.
Deck overhang forms for rolled beams spans, due to there
shallow depth, are typically supported in this manner.
B.6 - 36
September 2007
SPECIFICATIONS
COMMENTARY
6.10.3.2.4.3P Deck Slab Overhang Rotation
C6.10.3.2.4.3P
The designer shall consider the effects of out-of-plane girder
rotations, common with skewed bridges, on deck elevations.
Out-of-plane girder rotations will cause the overhang
formwork to also rotate. In an increasing magnitude from
the web of the fascia girder to the outside edge of the
formwork, the formwork will move upward or downward,
depending on the direction of rotation, during the deck pour.
It may be desirable to pre-rotate the overhang formwork so
that the as-designed deck overhang cross slope is obtained
after the deck pour is complete. Additionally, it may be
desirable to relocate the deck finishing machine support
railing from its typical position on the overhang formwork
to the fascia girders. This will minimize the upward and
downward movements of the finishing machine during the
deck pour due to out-of-plane girder rotations. Hand
finishing work will be necessary for the deck area beyond
the limits of the finishing machine.
The designer may consider approximating the anticipated
girder rotation based on girder differential vertical
displacements.
6.10.3.5 DEAD LOAD DEFLECTION
The provisions of A6.7.2 and D6.7.2 shall apply, as
applicable.
6.10.4 Service Limit State
6.10.4.2 PERMANENT DEFORMATION
6.10.4.2.1 General
C6.10.4.2.1
Delete the second paragraph of A6.10.4.2.1.
Delete the second paragraph of A6.10.4.2.1.
6.10.4.2.2 Flexure
C6.10.4.2.2
Delete the first paragraph following the first where list in
A6.10.4.2.2
Lateral bending in the bottom flange is only a
consideration at the service limit state for all horizontally
curved I-girder bridges and for straight I-girder bridges with
discontinuous cross-frame or diaphragm lines in conjunction
with skews less than 70 . Wind load and deck overhang
effects are not considered at the service limit state.
Delete the seventh paragraph of AC6.10.4.2.2.
B.6 - 37
September 2007
SPECIFICATIONS
COMMENTARY
6.10.5.1 FATIGUE
C6.10.5.1P
For horizontally curved I-girder bridges, the fatigue stress
range due to major-axis bending plus lateral bending shall
be investigated.
In horizontally curved I-girder bridges, the base metal
adjacent to butt welds and welded attachments on discretely
braced flanges subject to a net applied tensile stress must be
checked for the fatigue stress range due to major-axis
bending, plus flange lateral bending, at the critical
transverse location on the flange. Examples of welded
attachments for which this requirement applies include
transverse stiffeners and gusset plates receiving lateral
bracing members. The base metal adjacent to flange-to-web
welds need only be checked for the stress range due to
major-axis bending since the welds are located near the
center of the flange. Flange lateral bending need not be
considered for details attached to continuously braced
flanges.
6.10.5.3 SPECIAL FATIGUE REQUIREMENT FOR
WEBS
The following shall replace the first paragraph of A6.10.5.3
The live load flexural stress and shear stress resulting from
the fatigue load, as specified in A3.6.1.4 and D3.6.1.4, shall
be factored by two times the Pennsylvania Traffic Factor
(PTF) and the fatigue load factor specified for the fatigue
load combination in Table A3.4.1-1. The PTF is specified
in D6.6.1.2.2.
For the purposes of this article, the factored fatigue load
shall be taken as twice that calculated using the Fatigue load
combination specified in Tables A3.4.1-1 and in D3.4.1.1P,
with the fatigue live load taken as specified in A3.6.1.4 and
D3.6.1.4, multiplied by the Pennsylvania Traffic Factor
(PTF). The PTF is specified in D6.6.1.2.2.
6.10.6.2 FLEXURE
6.10.6.2.2 Composite Sections in Positive Flexure
C6.10.6.2.2
Composite sections in kinked (chorded) continuous or
horizontally curved steel girder bridges shall be considered
as noncompact sections and shall satisfy the requirements of
A6.10.7.2.
Composite sections in positive flexure in kinked (chorded)
continuous or horizontally curved steel bridges are also to
be designed at the strength limit state as noncompact
sections as specified in A6.10.7.2. Research has not yet been
conducted to support the design of these sections for a
nominal flexural resistance exceeding the moment at first
yield.
6.10.6.2.3 Composite Sections in Negative Flexure and
C6.10.6.2.3
B.6 - 38
September 2007
SPECIFICATIONS
COMMENTARY
Noncomposite Sections
Sections in kinked (chorded) continuous or horizontally
curved steel girder bridges shall be proportioned according
to provisions specified in A6.10.8 and D6.10.8.
For composite sections in negative flexure and
noncomposite sections, the provisions of A6.10.8 and
D6.10.8 limit the nominal flexural resistance to be less than
or equal to the moment at first yield. As a result, the
nominal flexural resistance for these sections is
conveniently expressed in terms of the elastically computed
flange stress.
For composite sections in negative flexure or
noncomposite sections in horizontally curved bridges, the
provisions of A6.10.8 and D6.10.8 must be used. Research
has not yet been conducted to extend the provisions of
Appendix A to sections in kinked (chorded) continuous or
horizontally curved steel bridges.
6.10.8 Flexural Resistance - Composite Sections in
Negative Flexure and Noncomposite Sections
6.10.8.2 COMPRESSION-FLANGE FLEXURAL
RESISTANCE
6.10.8.2.3 Lateral Torsional Buckling Resistance
The following shall replace Equation A6.10.8.2.3-10.
f1
=
stress without consideration of lateral bending at
the brace point opposite to the one corresponding
to f2, calculated as the intercept of the most critical
assumed linear stress variation passing through f2
and either fmid or f0 , whichever produces the
smaller value of Cb (MPa). f1 may be determined
as follows:
When the variation in the moment along the
entire length between the brace points is
concave in shape:
f1
f0
(6.10.8.2.3-10)
Otherwise:
f1
2 f mid
f2
f0
(6.10.8.2.3-11)
B.6 - 39
September 2007
SPECIFICATIONS
COMMENTARY
6.10.9 Shear Resistance
6.10.9.1 GENERAL
The following shall replace the first bulleted item of the
third paragraph of A6.10.9.1.
without a longitudinal stiffener and with a transverse
stiffener spacing not exceeding 1.5D, or
The following shall replace the fourth paragraph of
A6.10.9.1.
Provisions for end panels shall be as specified in
A6.10.9.3.3 and D6.10.9.3.3.
Transverse stiffener spacing shall also satisfy the
requirements of D6.10.3.2.5.2P for deck slab overhang form
support.
6.10.9.3.3 End Panels
A6.10.9.3.3.
The transverse stiffener spacing for end panels without a
longitudinal stiffener shall not exceed 0.5D. The transverse
stiffener spacing of end panels with a longitudinal stiffener
shall not exceed 0.5 times the maximum subpanel depth.
6.10.10 Shear Connectors
6.10.10.1 GENERAL
C6.10.10.1P
A6.10.10.1.
Shear connectors are required along the entire length of the
girder when a composite girder analysis has been
performed.
Mechanical shear connectors provide for the horizontal
shear at the interface between the concrete slab and the steel
girder in the positive moment regions and the horizontal
shear between the longitudinal reinforcement steel within
the effective flange width and the steel girder in the negative
moment regions.
6.10.10.1.1 Types
The minimum diameter of studs shall be 19 mm {3/4 in.}.
6.10.10.1.2 Pitch
C6.10.10.1.2
The following shall replace the part of the article starting
immediately after Equation A6.10.10.1.2-1 to the end of
A6.10.10.1.2.
in which:
At the fatigue limit state, shear connectors are designed
for the range of live load shear between the deck and top
flange of the girder. In straight girders, the shear range
normally is due to only major-axis bending if torsion is
ignored. Curvature, skew and other conditions may cause
torsion, which introduces a radial component of the
Vsr =
horizontal fatigue shear range per unit length
(N/mm) {kip-in.}
B.6 - 40
September 2007
SPECIFICATIONS
=
V fat
2
+ Ffat
COMMENTARY
2
(6.10.10.1.2-2)
Vfat =
longitudinal fatigue shear range per unit length
(N/mm) {kip/in.}
Vf Q
=
(6.10.10.1.2-3)
I
Ffat =
radial fatigue shear range per unit length (N/mm)
{kip/in} taken as the larger of either:
Ffat1 =
Abot σ flg
wR
(6.10.10.1.2-4)
or
Ffat2 =
Frc
w
(6.10.10.1.2-5)
where:
flg
=
range of longitudinal fatigue stress in the bottom
flange without consideration of flange lateral
bending (MPa) {ksi}
Abot =
area of the bottom flange (mm2) {in.2}
Frc =
net range of cross-frame or diaphragm force at the
top flange (N) {kip}
I
=
moment of inertia of the short-term composite
section (mm4) {in.4}

=
distance between brace points (mm) {ft.}
n
=
number of shear connectors in a cross-section
p
=
pitch of shear connectors along the longitudinal
axis (mm) {in.}
Q
=
first moment of the transformed short-term area of
the concrete deck about the neutral axis of the
short-term composite section (mm3) {in.3}
R
=
minimum girder radius within the panel (mm) {ft.}
Vf =
vertical shear force range under the fatigue load
combination specified in Table A3.4.1-1 and
D3.4.1-1 with the fatigue live load taken as
specified in A3.6.1.4 and D3.6.1.4 (N) {kip}
w
effective length of deck (mm) taken as 1220 mm
{48 in.}, except at end supports where w may be
=
horizontal shear. These provisions provide for consideration
of both of the components of the shear to be added
vectorially according to Eq. 2.
The radial shear range, Ffat, typically is determined for
the fatigue live load positioned to produce the largest
positive and negative major-axis bending moments in the
span. Therefore, vectorial addition of the longitudinal and
radial components of the shear range is conservative
because the longitudinal and radial shears are not produced
by concurrent loads.
Eq. 4 may be used to determine the radial fatigue shear
range resulting from the effect of any curvature between
brace points. The shear range is taken as the radial
component of the maximum longitudinal range of force in
the bottom flange between brace points, which is used as a
measure of the major-axis bending moment. The radial
shear range is distributed over an effective length of girder
flange, w. At end supports, w is halved. Eq. 4 gives the
same units as Vfat.
Eq. 5 will typically govern the radial fatigue shear
range where torsion is caused by effects other than
curvature, such as skew. Eq. 5 is most likely to control only
in regions adjacent to a skewed support for which the skew
angle less than 70° in either a straight or horizontally curved
bridge. Eqs. 4 and 5 yield approximately the same value if
the span or segment is curved and there are no other sources
of torsion in the region under consideration. Note that Frc
represents the net range of force transferred to the top flange
from all cross-frames at the point under consideration due to
the factored fatigue load plus impact. In lieu of a refined
analysis, Frc for an exterior girder, which is the critical
case, may be taken as 111 200 N {25 kips}, and taken as
zero for interior girders. Eq. 5 should only be checked using
this value within the regions of a straight or horizontally
curved girder in the region of a skewed support with a skew
angle less than 70°. Regardless of whether Frc is determined
by refined analysis or taken as the above recommended
value, it should be multiplied by the factor of 0.75 discussed
in Article C6.6.1.2.1 and DC6.6.1.2.1 to account for the
probability of two vehicles being in their critical relative
position to cause the maximum range of cross-frame force at
the top flange.
Eqs. 4 and 5 are provided to ensure that a load path is
provided through the shear connectors to satisfy equilibrium
at a transverse section through the girders, deck and crossframe.
B.6 - 41
September 2007
SPECIFICATIONS
COMMENTARY
taken as 610 mm {24 in.}
Zr =
shear fatigue resistance of an individual shear
connector determined as specified in A6.10.10.2
and D6.10.10.2 (N) {kip}
For straight spans or segments, the radial fatigue shear range
from Eq. 4 may be taken equal to zero. For straight or
horizontally curved bridges with skews not less than 70°,
the radial fatigue shear range from Eq. 5 may be taken equal
to zero.
The center-to-center pitch of shear connectors shall not
exceed 600 mm {24 in.} and shall not be less than six stud
diameters.
The center-to-center pitch of channel shear connectors shall
not exceed 600 mm {24 in.} and shall not be less than
150 mm {6 in.}.
6.10.10.1.3 Transverse Spacing
The minimum number of studs in a group shall consist of
two in a single transverse row.
6.10.10.1.4 Cover and Penetration
C6.10.10.1.4
The clear depth of concrete cover over the tops of the shear
connectors should not be less than 60 mm {2 1/2 in.}.
Shear connectors should penetrate at least 50 mm {2 in.}
into the deck.
Delete the second sentence of AC6.10.10.1.4.
For plan presentation, show cover and penetration limits; do
not detail stud height (see BC-753M). Stud heights are
determined in the field based on actual girder elevations.
6.10.10.1.5P SPLICE LOCATIONS
Shear connectors at splice locations shall be arranged to
clear fasteners and shall be welded to the splice plate. Up to
20% fewer connectors, than required by design, are
acceptable in the splice zone, provided that the deleted
connectors are furnished as additional connectors adjacent
to the splice.
6.10.10.2 FATIGUE RESISTANCE
C6.10.10.2
The fatigue resistance of an individual channel shear
connector, Zr, shall be taken as:
Metric Units:
Zr = B w
184 w
(6.10.10.2-3P)
Equations 3P and 4P were added because the LRFD
specification does not address the fatigue resistance of
channel shear connectors. Equations 3P and 4P were
converted from Article 10.3.8.5.1.1 of AASHTO Standard
Specification for Highway Bridges.
B.6 - 42
September 2007
SPECIFICATIONS
Zr = B w
COMMENTARY
1.05 w
where:
Metric Units:
(6.10.10.2-4P)
B = 1525 - 175 Log N
B = 8.7 - Log N
w
=
length of channel shear connector measured in a
transverse direction on the flange of a girder (mm)
N
=
number of cycles specified in A6.6.1.2.5 and
D6.6.1.2.5
6.10.10.3 SPECIAL REQUIREMENTS FOR POINTS OF
PERMANENT LOAD CONTRAFLEXURE
C6.10.10.3
Delete A6.10.10.3.
Delete AC6.10.10.3.
PennDOT requires composite girders to be composite fulllength of the bridge with shear connectors full-length of the
bridge.
6.10.10.4 STRENGTH LIMIT STATE
6.10.10.4.2 Nominal Shear Force
6.10.10.4.2
For simple spans and for continuous spans that are
noncomposite for negative flexure in the final condition, the
total nominal shear force, P, between the point of maximum
positive design live load plus impact moment and each
adjacent point of zero moment shall be taken as:
The radial effect of curvature is included in Eqs. 4 and 9.
For curved spans or segments, the radial force is required to
bring into equilibrium the smallest of the longitudinal forces
in either the deck or the girder. When computing the radial
component, the longitudinal force is conservatively assumed
to be constant over the entire length Lp or Ln, as applicable.
P
Pp 2
Fp 2
(6.10.10.4.2-1)
in which:
Pp =
total longitudinal shear force in the concrete deck
at the point of maximum positive live load plus
impact moment (N) {kip} taken as the lesser of
either:
P1 p
(6.10.10.4.2-2)
0.85 fc bs ts
or
P2 p
Fyw Dtw
Fyt bft t ft
Fycbfct fc
(6.10.10.4.2-3)
B.6 - 43
September 2007
SPECIFICATIONS
Fp =
COMMENTARY
total radial shear force in the concrete deck at the
point of maximum positive live load plus impact
moment (N) {kip} taken as:
Fp
Pp
Lp
(6.10.10.4.2-4)
R
where:
bs
=
effective width of the concrete deck (mm)
Lp =
arc length between an end of the girder and an
adjacent point of maximum positive live load plus
impact moment (mm) {ft}
R
=
minimum girder radius over the length, Lp (mm)
{ft.}
ts
=
thickness of the concrete deck (mm) {in.}
For straight segments, Fp may be taken equal to zero.
For continuous spans that are composite for negative flexure
in the final condition, the total nominal shear force, P,
between the point of maximum positive design live load
plus impact moment and an adjacent end of the member
shall be determined from Eq. 1. The total nominal shear
force, P, between the point of maximum positive design live
load plus impact moment and the centerline of an adjacent
interior support shall be taken as:
P
PT 2
FT 2
(6.10.10.4.2-5)
in which:
PT =
between the point of maximum positive live load
plus impact moment and the centerline of an
adjacent interior support (N) {ft.} taken as:
PT
Pn =
Pp
(6.10.10.4.2-6)
Pn
over an interior support (N) {ft.} taken as the lesser
of either:
P1n
Fyw Dtw
Fyt bft t ft
Fycbfct fc
(6.10.10.4.2-7)
or
P2n
0.45 f c bs ts
(6.10.10.4.2-8)
B.6 - 44
September 2007
SPECIFICATIONS
FT =
COMMENTARY
total radial shear force in the concrete deck
between the point of maximum positive live load
plus impact moment and the centerline of an
adjacent interior support (N) {kip} taken as:
FT
PT
Ln
R
(6.10.10.4.2-9)
where:
Ln =
arc length between the point of maximum positive
live load plus impact moment and the centerline of
an adjacent interior support (mm) {ft.}
R
minimum girder radius over the length, Ln (mm)
{ft.}
=
For straight segments, FT may be taken equal to zero.
6.10.11 Stiffeners
6.10.11.1 TRANSVERSE INTERMEDIATE
STIFFENERS
6.10.11.1.1 General
C6.10.11.1.1P
Single-sided stiffeners on horizontally curved girders should
be attached to both flanges. When pairs of transverse
stiffeners are used on horizontally curved girders, they shall
be fitted tightly to both flanges.
Transverse stiffeners shall also satisfy the requirements
given in Standard Drawing BC-753M.
A6.10.11.1.1.
The distance between the end of the web-to-stiffener weld
and the near edge of the adjacent web-to-flange or
longitudinal stiffener-to-web weld shall not be less than 4tw
but not to exceed the lesser of 6tw and 100 mm {4 in.}.
When single-sided transverse stiffeners are used on
horizontally curved girders, they should be attached to both
flanges to help to restrain the flanges and to help retain the
cross-sectional configuration of the girder when subjected to
torsion. The fitting of pairs of transverse stiffeners against
the flanges is required for the same reason.
The minimum distance between the end of the web-tostiffener weld to the adjacent web-to-flange or longitudinal
stiffener-to-web weld is set to relieve flexing of the
unsupported segment of the web to avoid fatigue-induced
cracking of the stiffener-to-web welds, and to avoid
inadvertent intersecting welds. The 6tw criterion for
maximum distance is set to avoid vertical buckling of the
unsupported web. The 100 mm {4 in.} criterion was
arbitrarily selected to avoid a very large unsupported length
where the web thickness has been selected for reasons other
than stability, e.g., webs of bascule girders at trunnions.
6.10.11.1.2 Projecting Width
For tub sections in regions of negative flexure, bf shall be
taken as the full-width of the widest top flange within the
field section under consideration.
6.10.11.3 LONGITUDINAL STIFFENERS
B.6 - 45
September 2007
SPECIFICATIONS
COMMENTARY
6.10.11.3.3 Moment of Inertia and Radius of Gyration
C6.10.11.3.3
Longitudinal stiffeners shall satisfy:
The rigidity required of longitudinal stiffeners on
curved webs is greater than the rigidity required on straight
webs because of the tendency of curved webs to bow. The
factor in Eq. 1 is a simplification of the requirement in the
Hanshin (1988) provisions for longitudinal stiffeners used
on curved girders. For longitudinal stiffeners on straight
webs, Eq. 5 leads to = 1.0.
I
Dtw3 2.4
do
D
2
0.13
(6.10.11.3.3-1)
and
Fys
0.16d o
E
Fyc
r
1 0.6
(6.10.11.3.3-2)
Rh Fys
in which:
=
curvature correction factor for longitudinal
stiffener rigidity calculated as follows:
For cases where the longitudinal stiffener is on the
side of the web away from the center of curvature:
Z
1
(6.10.11.3.3-3)
6
For cases where the longitudinal stiffener is on the
side of the web toward the center of curvature:
Z
1
(6.10.11.3.3-4)
12
Z
=
curvature parameter:
=
0.95do2
Rtw
10
(6.10.11.3.3-5)
where:
do =
transverse stiffener spacing (mm) {in.}
I
=
moment of inertia of the longitudinal stiffener
including an effective width of the web equal to
18tw taken about the neutral axis of the combined
section (mm4) {in.4}. If Fyw is smaller than Fys, the
strip of the web included in the effective section
shall be reduced by the ratio Fyw/Fys.
R
=
minimum girder radius in the panel (mm) {in.}
r
=
radius of gyration of the longitudinal stiffener
including an effective width of the web equal to
18tw taken about the neutral axis of the combined
B.6 - 46
September 2007
SPECIFICATIONS
COMMENTARY
section (mm) {in.}. If Fyw is smaller than Fys, the
strip of the web included in the effective section
shall be reduced by the ratio Fyw/Fys.
6.10.11.4P STIFFENERS IN RIGID-FRAME KNEES
6.10.11.4.1P Stiffener Spacing
The spacing of stiffeners in rigid-frame knees shall satisfy
both of the following equations:
fa
F yc - f cs
(6.10.11.4.1P-1)
fb
F yc
(6.10.11.4.1P-2)
and
for which:
f a = f cs
f b = f cs
=
3b 2
Rt
3b 2
Rt
2
(6.10.11.4.1P-3)
4
4+1.14
3
3.2+
(6.10.11.4.1P-4)
3
a
b
(6.10.11.4.1P-5)
where:
a
=
spacing of stiffness (mm) {in.}
b
=
half of flange width (mm) {in.}
Fyc =
specified minimum yield strength of a compression
flange (MPa) {ksi}
fcs =
maximum compression Service I load flange stress
(MPa) {ksi}
R
=
radius of flange curvature (mm) {in.}
t
=
thickness of flange (mm) {in.}
B.6 - 47
September 2007
SPECIFICATIONS
COMMENTARY
Figure 6.10.11.4.1P-1 - Stiffeners in Rigid-Frame Knees
6.10.11.4.2P Stiffener Design
The factored bearing resistance of stiffeners in rigid-frame
knees, taken as specified in A6.10.11.2.3, shall be greater
than Pb, taken as:
P b = f r ab
(6.10.11.4.2P-1)
for which:
fr =
f ct
R
(6.10.11.4.2P-2)
where:
a
=
spacing of stiffeners (mm) {in.}
b
=
half of flange width (mm) {in.}
fc
=
maximum factored compression flange stress
(MPa) {ksi}
R
=
radius of flange curvature (mm) {in.}
t
=
thickness of flange (mm) {in.}
B.6 - 48
September 2007
SPECIFICATIONS
COMMENTARY
6.10.12 Cover Plates
6.10.12.3P COVER PLATE LENGTH AND WIDTH
C6.10.12.3P
The length of any welded cover plate added to a rolled beam
shall extend the full-length of the rolled beam, including the
bearing area, or the full-length of the rolled beam field
section in the case of a spliced beam unless otherwise
approved by the Chief Bridge Engineer. The use of partial
length cover plates is allowed for rehabilitation projects
with detailed fatigue analysis. Partial length cover plates
must be a bolted connection at the ends.
The width of the plate shall not exceed the width of the
flange by 150 mm {6 in.}, or six times the thickness of the
cover plate, whichever is less. Bottom flange cover plates
preferably shall be wider than the bottom flange. Top
flange cover plates shall be of constant width, preferably
narrower than the top flange. When a cover plate narrower
than the flange is used, the width of the plate shall be at
least 50 mm {2 in.} less than the width of the flange. The
width of a cover plate connected by fillet welds shall be no
greater than 24 times the plate thickness.
The Department does not allow partial length cover plates
for new designs.
6.11 BOX SECTIONS IN FLEXURE
C6.11P
The provisions of A6.11 and D6.11 apply to steel box Igirders curved in plan. The provisions of Appendix E,
DE6.11P apply to straight box-girder bridges.
The provisions of D6.11 are based on the 2004 Third
Edition of AASHTO-LRFD Specifications as amended
herein. The provisions of Appendix E are based on the
1998 Second Edition of the AASHTO-LRFD Design
Specifications with the 1999 through 2003 Interims.
6.11.1 General
C6.11.1
The following shall replace the first sentence of the first
paragraph of AC6.11.1.
A6.11.1 and D6.11.1address general topics that apply to
closed-box and tub sections used as flexural members in
horizontally curved bridges, or bridges containing both
straight and curved segments. For the application of the
provisions of A6.11 and D6.11, bridges containing both
straight and curved segments are to be treated as
horizontally curved bridges since the effects of curvature on
the support reactions and girder deflections, as well as the
effects of flange lateral bending and torsional shear, usually
extend beyond the curved segments.
For horizontally curved boxes, flange lateral bending effects
due to curvature and the effects of torsional shear must
always be considered at all limit states and also during
construction.
B.6 - 49
September 2007
SPECIFICATIONS
COMMENTARY
6.11.1.1 STRESS DETERMINATION
C6.11.1.1
The following shall replace the third and fourth paragraphs
of A6.11.1.1.
The section of an exterior member assumed to resist
horizontal factored wind loading within these bridges may
be taken as the bottom box flange acting as a web and 12
times the thickness of the web acting as flanges.
The provisions of A4.6.2.2.2b and D4.6.2.2.2b shall not
apply to single or multiple box sections in horizontally
curved bridges. For these sections, the effects of both
flexural and St. Venant torsional shear shall be considered.
The St. Venant torsional shear stress in box flanges due to
the factored loads at the strength limit state shall not exceed
the factored torsional shear resistance of the flange, F vr,
taken as:
Fyf
Fvr 0.75 v
(6.11.1.1-1)
3
Delete the seventh paragraph of AC6.11.1.1
6.11.3.2 FLEXURE
C6.11.3.2
The following shall replace the first four sentences of the
third paragraph of AC6.11.3.2.
For horizontally curved girders, flange lateral bending
effects due to curvature must always be considered during
construction.
Delete the fourth paragraph of AC6.11.3.2.
fourth paragraph of A6.11.5.
Longitudinal warping stresses and transverse bending
stresses due to cross-section distortion shall be considered
for single and multiple box sections in horizontally curved
bridges.
The following shall supplement the last paragraph of
A6.11.5.
Unless adequate strength and stability of a damaged
structure can be verified by refined analysis, in crosssections comprised of two box sections, only the bottom
flanges in the positive moment regions should be designated
as fracture-critical. Where cross-sections contain more than
two box girder sections, none of the components of the box
sections should be considered fracture-critical.
B.6 - 50
September 2007
SPECIFICATIONS
COMMENTARY
6.11.6.2 FLEXURE
6.11.6.2.2 Sections in Positive Flexure
C6.11.6.2.2
Sections in horizontally curved steel girder bridges shall be
considered as noncompact sections and shall satisfy the
requirements of A6.11.7.2 and A6.10.7.3.
For sections in positive flexure in horizontally curved
bridges the nominal flexural resistance is not permitted to
exceed the moment at first yield. The nominal flexural
resistance in these cases is therefore more appropriately
expressed in terms of the elastically computed flange stress.
If the section is part of a horizontally curved bridge, the
section must be designed as a noncompact section. The
ability of such sections to develop a nominal flexural
resistance greater than the moment at first yield in the
presence of potentially significant St. Venant torsional shear
and cross-sectional distortion stresses has not been
demonstrated.
Noncompact sections must also satisfy the ductility
requirement specified in 6.10.7.3 to ensure a ductile failure.
Satisfaction of this requirement ensures an adequate margin
of safety against premature crushing of the concrete deck for
sections utilizing 690-MPa {100 ksi} steels and/or for
sections utilized in shored construction. This requirement is
also a key limit in allowing web bend-buckling to be
disregarded in the design of composite sections in positive
flexure when the web also satisfies A6.11.2.1.2, as
discussed in AC6.10.1.9.1.
6.11.9 Shear Resistance
The following shall replace the third paragraph of A6.11.9.
For all horizontally curved sections, Vu shall be taken as the
sum of the flexural and St. Venant torsional shears.
6.11.10 Shear Connectors
6.11.10
The following shall replace the third paragraph of A6.11.10.
For horizontally curved sections, shear connectors shall
be designed for the sum of the flexural and St. Venant
torsional shears. The longitudinal fatigue shear range per
unit length, Vfat, for one top flange of a tub girder shall be
computed for the web subjected to additive flexural and
torsional shears. The resulting shear connector pitch shall
also be used for the other top flange. The radial fatigue
shear range due to curvature, Ffat1, given by
Eq. D6.10.10.1.2-4 may be ignored in the design of box
sections in straight or horizontally curved spans or
segments.
For checking the resulting number of shear connectors
to satisfy the strength limit state, the cross-sectional area of
Delete the second sentence of the first paragraph of
A6.11.10.
The following shall supplement the second paragraph of
A6.11.10.
Because of the inherent conservatism of these
requirements, the radial fatigue shear range due to curvature
need not be included when computing the horizontal fatigue
shear range for box sections in either straight or horizontally
curved spans or segments.
B.6 - 51
September 2007
SPECIFICATIONS
COMMENTARY
the steel box section under consideration and the effective
area of the concrete deck associated with that box shall be
used in determining P by Eqs. D6.10.10.4.2-2,
D6.10.10.4.2-3, D6.10.10.4.2-7, and D6.10.10.4.2-8.
The following shall replace the first sentence of last
paragraph of A6.11.10.
For composite box flanges at the fatigue limit state, Vsr
in Eq. A6.10.10.1.2-1 shall be determined as the vector sum
of the longitudinal fatigue shear range given by
Eq. D6.10.10.1.2-3 and the torsional fatigue shear range in
the concrete deck
6.11.11 Stiffeners
6.11.11.2 LONGITUDINAL COMPRESSION-FLANGE
STIFFENERS
C6.11.11.2
A6.11.11.2.
For structural tees, b should be taken as one-half the width
of the flange.
6.12 MISCELLANEOUS FLEXURAL MEMBERS
6.12.1 General
6.12.1.2 STRENGTH LIMIT STATE
6.12.1.2.3 Shear
The following shall replace the definition of n in
A6.12.1.2.3.
Vn =
nominal shear resistance specified in A6.10.9.2 and
DE6.10.7.2P, as appropriate, for webs of
noncomposite members and 6.12.3 for webs of
composite members (N) {kip}
6.12.2 Nominal Flexural Resistance
6.12.2.2 NONCOMPOSITE MEMBERS
6.12.2.2.1 I- and H-Shaped Members
A6.12.2.2.1.
The provisions of A6.10, D6.10, DE6.10P and DE6.11p
shall apply to flexure about an axis perpendicular to the
web.
B.6 - 52
September 2007
SPECIFICATIONS
COMMENTARY
6.13 CONNECTIONS AND SPLICES
6.13.1 General
C6.13.1
The following shall replace the first paragraph of A6.13.1.
Except as specified otherwise, connections and splices for
main members (both flanges and webs) shall be designed at
the strength limit state for not less than the larger of:
The average of the flexural moment-induced stress,
shear, or axial force due to the factored loadings at the
point of splice or connection and the factored flexural,
shear, or axial resistance of the member or element at
the same point, or
Often stresses on the top flange of a composite girder are
low under design loads. This criteria would require a top
splice plate to be about 75% of the area of the top flange,
and have the appropriate number of bolts.
75 percent of the factored flexural, shear, or axial
resistance of the member or element.
If it is necessary to cope a flange in order to provide
clearance at the end connection of a floorbeam or stringer,
the bending resistance of the member at the cope location
shall not be decreased by more than 50%. No sharp notches
shall be introduced as a result of coping. The maximum
practical radius shall be maintained at all copes with an
absolute minimum radius of 50 mm {2 in.}.
Where diaphragms, cross-frames, lateral bracing,
stringers, or floorbeams for straight or horizontally curved
flexural members are included in the structural model used
to determine force effects, or alternatively, are designed for
explicitly calculated force effects from the results of a
separate investigation, end connections for these bracing
members shall be designed for the calculated factored
member force effects. Otherwise, the end connections for
these members shall be designed according to the 75 percent
resistance provision contained herein.
The exception for bracing members for straight or
horizontally curved flexural members, that are included in
the structural model used to determine force effects, results
from experience with details developed invoking the
75 percent and average load provisions herein. These details
tended to become so large as to be unwieldy resulting in
large eccentricities and force concentrations. It has been
decided that the negatives associated with these connections
justifies the exception permitted herein.
6.13.2 Bolted Connections
6.13.2.1 GENERAL
C6.13.2.1P
A6.13.2.1.
High-strength bolted joints shall be designated as slipcritical connections. Bearing-type connections may be used
on rehabilitation projects if approved by the Chief Bridge
Engineer.
When detailing bolted connections, tightening clearance
between flange and web bolts need to be taken into account.
Manual of Steel Construction, American Institute of Steel
Construction, provides information on assembling
clearances for threaded fasteners which can be used to avoid
bolt interference problems.
B.6 - 53
September 2007
SPECIFICATIONS
COMMENTARY
6.13.2.3 BOLTS, NUTS AND WASHERS
6.13.2.3.2 Washers
The following shall replace the third and fifth bulleted items
of A6.13.2.3.2.
AASHTO M 253 (ASTM A 490) bolts are to be
installed in material having a specified minimum yield
strength less than 345 MPa {50 ksi}, irrespective of the
tightening method;
AASHTO M 253 (ASTM A 490) bolts over 25.4 mm
{1 in.} in diameter are to be installed in an oversize or
short-slotted hole in an outer-ply, in which case a
minimum thickness of 7.9 mm {5/16 in.} shall be used
under both the head and the nut. Multiple hardened
washers shall not be used.
6.13.2.4 HOLES
6.13.2.4.1 Types
6.13.2.4.1b Oversize Holes
The following shall replace A6.13.2.4.1b.
Approval of Chief Bridge Engineer must be obtained before
oversize holes can be used in any or all plies of slip-critical
connections. Oversize holes are not permitted in
diaphragms or cross frames of curved girder bridges.
Oversize holes shall not be used in bearing-type
connections.
C6.13.2.4.1c Short-Slotted Holes
On skew bridges, short slotted holes for cross-frame
connections should be utilized for adjustment for temporary
and permanent conditions.
6.13.2.4.2 Size
B.6 - 54
September 2007
SPECIFICATIONS
COMMENTARY
Table 6.13.2.4.2-1 - Maximum Hole Sizes
Metric Units
Bolt
Diameter
d
(mm)
Standard
Hole
Diameter
(mm)
Oversize
Hole
Diameter
(mm)
Short Slot
Width x Length
(mm x mm)
Long Slot
Width x Length
(mm x mm)
15.9
17.5
20.7
17.5 x 22
17.5 x 40
19.1
20.7
23.8
20.7 x 25
20.7 x 48
22.2
23.8
27.0
23.8 x 29
23.8 x 56
25.4
28.6
27.0
31.8
27.0 x 33
27.0 x 64
d+1.6
d+7.9
d+1.6 x d+9.5
d+1.6 x 2.5d
Bolt
Diameter
d
(in.)
Standard
Hole
Diameter
(in.)
Oversize
Hole
Diameter
(in.)
Short Slot
Width x Length
(in. x in.)
Long Slot
Width x Length
(in. x in.)
5/8
11/16
13/16
11/16 x 7/8
11/16 x 1-9/16
3/4
13/16
15/16
13/16 x 1
13/16 x 1-7/8
7/8
15/16
1-1/16
15/16 x 1-1/8
15/16 x 2-3/16
1
1-1/16
1-1/4
1-1/16 x 1-5/16
1-1/16 x 2-1/2
d+1/16
d+5/16
d+1/16 x d+3/8
d+1/16 x 2.5d
1-1/8
6.13.2.5 SIZE OF BOLTS
C6.13.2.5P
Bolts shall not be less than 15.9 mm {5/8 in.} in diameter.
Bolts 15.9 mm {5/8 in.} in diameter shall not be used in
primary members, except in 64 mm {2.5 in.} legs of angles
and in flanges of sections whose dimensions require 15.9
mm {5/8 in.} fasteners to satisfy other detailing provisions
herein. Use of structural shapes which do not allow the use
of 15.9 mm {5/8 in.} fasteners shall be limited to handrails.
Angles whose size is not determined by a calculated
demand may use:
Typically, high-strength bolts will be 22.2 mm {7/8 in.}
diameter mechanically galvanized AASHTO M 164 (ASTM
A 325) bolts. This is the typical high-strength bolt used in
the past.
15.9 mm {5/8 in.} diameter bolts in 51 mm {2 in.} legs,
19.1 mm {3/4 in.} diameter bolts in 64 mm {2 1/2 in.}
legs,
22.2 mm {7/8 in.} diameter bolts in 76 mm {3 in.} legs,
and
25.4 mm {1 in.} diameter bolts in 89 mm {3 1/2 in.}
legs.
B.6 - 55
September 2007
SPECIFICATIONS
COMMENTARY
The diameter of bolts in angles of primary members shall
not exceed one-fourth the width of the leg in which they are
placed.
Fasteners shall be of the size shown on the contract plans,
but generally shall be 22.2 mm {7/8 in.} in diameter.
6.13.2.6 SPACING OF BOLTS
C6.13.2.6.1 Minimum Spacing and Clear Distance
The preferred distance between centers of bolts in standard
holes shall not be less than the values in Table 1:
Table C6.13.2.6.1-1 - Preferred Bolt Spacing
Metric Units
Diameter
Bolt
(mm)
Preferred
Distance between
Centers
of Bolts
(mm)
Diameter
Bolt
(in.)
Preferred
Distance between
Centers
of Bolts
(in.)
15.9
60
5/8
2 1/4
19.1
65
3/4
2 1/2
22.2
75
7/8
3
25.4
90
1
3 1/2
6.13.2.6.6 Edge Distances
B.6 - 56
September 2007
SPECIFICATIONS
COMMENTARY
Table 6.13.2.6.6-1 - Minimum Edge Distances
Metric Units
Bolt
Diameter
(mm)
Sheared or
Gas Cut
Edges (mm)
Rolled Edges
of Plates or
Shapes (mm)
Bolt
Diameter
(in.)
Sheared or
Gas Cut
Edges (in.)
Rolled Edges
of Plates or
Shapes (in.)
15.9
29
22
5/8
1 1/8
7/8
19.1
32
25
3/4
1 1/4
1
22.2
38
29
7/8
1 1/2
1 1/8
25.4
44
32
1
1 3/4
1 1/4
28.6
51
38
1 1/8
2
1 1/2
31.8
57
41
1 1/4
2 1/4
1 5/8
34.9
60
44
1 3/8
2 3/8
1 3/4
6.13.2.7 SHEAR RESISTANCE
C6.13.2.7
AC6.13.2.7.
For steel plate girder flange splices greater than 1270 mm
{50 in.} in length, the 20% reduction shall be applied to the
nominal shear resistance of a bolt, calculated using Equation
1 and 2, because the axial force is parallel to the line of
bolts. In such flange splices, the 1270 mm {50 in.} length is
to be measured between the extreme bolts on only one side
of the connection. The 20 percent reduction should not be
applied for web bolts subjected to shear and moment.
6.13.2.8 SLIP RESISTANCE
C6.13.2.8
B.6 - 57
September 2007
SPECIFICATIONS
COMMENTARY
Table 6.13.2.8-1 - Minimum Required Bolt Tension
Metric Units
Required TensionPt (kN)
Bolt
Diametert
(mm)
M 164
(A 325)
15.9
19.1
Required TensionPt(kip)
M 253
(A490)
Bolt
Diameter
(in.)
M 164
(A 325)
M 253
(A 490)
84.5
120
5/8
19
24
125
178
3/4
28
35
22.2
173
245
7/8
39
49
25.4
227
325
1
51
64
28.6
249
409
1 1/8
56
80
31.8
320
516
1 1/4
71
102
34.9
378
618
1 3/8
85
121
38.1
463
752
1 1/2
103
148
The following shall replace the first bulleted item in the
second paragraph (the definition of Class A surface).
!
Class A surface: blast cleaned surfaces with Class A
coatings
For values of Ks, use Class A surface conditions for
design, unless a paint is tested and proven to conform to
Class B conditions. If Class B is used, field testing and
controls must be specified in the contract drawings or
construction specifications.
Delete the fourth paragraph in A6.13.2.8 which starts
"The contract document shall specify that joints
having..."and replace it with the following.
The following note shall be placed on the contact
drawings:
“Blast clean the faying surfaces of splices and
connections of all structural elements in accordance
with Publication 408 Section 1060.3(b)3. Reblast
unpainted elements that remain unassembled for a
period of 12 months or more following the initial
cleaning.”
The revision to the definition of Class A and the
requirement to blast clean all faying surfaces is based on
results of research conducted jointly by the University of
Texas at Austin and the FHWA in the early 1980's on
weathering steel connections. An extensive testing
program conducted in conjunction with the research
showed that weathering steel connections with a mill scale
surface had an average slip coefficient, Ks, less than the
0.33 value for Class A. Blast cleaned weathering steel
achieved an average slip coefficient above the 0.50 value
specified for a Class B contact surface. The testing
program incorporated a wide range of variables, including
exposure of test specimens to an open environment for
periods up to 12 months.
The UT/FHWA research suggests that present LRFD
design policy, which allows a mill scale surface for Class
A, could result in weathering steel connections that do not
meet the slip coefficient value for a Class A contact
surface. The revision to the definition of Class A and the
requirements to blast clean all faying surfaces will add
desired safety into Department projects.
Inherent factors of safety in the design of connections
should ensure the serviceability of in-place weathering
steel structures where the slip critical condition controlled
the design.
Designers are directed to review Publication 35 Bulletin
15 for current paint systems and corresponding slip
coefficients.
6.13.2.10 Tensile Resistance
6.13.2.10.3 Fatigue Resistance
C6.13.2.10.3
B.6 - 58
SPECIFICATIONS
September 2007
COMMENTARY
Replace references to A 325M and A 490M with A 325
6.13.2.11 COMBINED TENSION AND SHEAR
C6.13.2.11
Replace references to A 325M with A 325.
6.13.3 Welded Connections
6.13.3.1 GENERAL
C6.13.3.1
Field welding is generally prohibited. Provisions may be
made for attachment of stay-in-place forms, bearing plates
and sole plates of pot bearings (but not the pot bearing
itself). All areas where field welding is permitted shall be
specifically designated on the contract plans. The fatigue
provisions of this specification shall apply to the design of
all affected members.
The regions of welded structures requiring nondestructive testing (NDT), along with the allowable types
of NDT, shall be shown on the contract plans.
The following shall supplement AC6.13.3.1
The AASHTO/AWS D1.5M/D1.5:2002 Bridge Welding
Code describes the appropriate application of the types of
NDT.
Use the AASHTO/AWS D1.1M/D1.1:2002 Structural
Welding Code for the welding of new tubular structures,
pipes, piles and existing steel which are not covered by
AASHTO/AWS D1.5M/D1.5:2002.
6.13.3.8P INTERSECTING WELDS
Intersecting welds which provide a potential crack path
into the web or flange of a girder from an attachment will
not be permitted. The termination of the fillet weld to
prevent the intersection shall provide a minimum
clearance of 40 mm {1 1/2 in.}, unless another clearance
is required by other design documents. Transverse groove
welds shall not be terminated to prevent the intersection.
B.6 - 59
SPECIFICATIONS
September 2007
COMMENTARY
6.13.3.9P INTERMITTENT FILLET WELDS
Intermittent fillet welds are prohibited, unless they are
incorporated in the final weld in accordance with
AASHTO/AWS Bridge Welding Code.
6.13.3.10P MINIMUM EDGE DISTANCE
C6.13.3.10P
A minimum edge distance of 25 mm {1 in.} shall be
maintained from a fillet weld termination to the edge of a
base metal plate in the direction of the weld.
An example of minimum edge distance is graphically
shown in Figure C1.
Figure C6.13.3.10P-1 - Minimum Edge Distance
6.13.4 Block Shear Rupture Resistance
The following shall replace the second sentence of the
fourth paragraph of A6.13.4.
The net area shall be the gross area, minus the number of
holes or fractional holes in the plane, multiplied by the
nominal hole diameter specified in Table D6.13.2.4.2-1
plus 1.6 mm {1/16 in.} times the thickness of the
component.
6.13.6 Splices
6.13.6.1 BOLTED SPLICES
6.13.6.1.1 General
Bolted splices shall be designed at the strength limit state
to satisfy the requirements specified in A6.13.1 and
D6.13.1. Where a section changes at a splice, the smaller
of the two connected sections shall be used in the design.
Develop bolted field splices for steel beams and girders in
accordance with BD-616M. Splices shall be designed
using the Department’s SPLRFD program.
6.13.6.1.4 Flexural Members
C6.13.6.1.4P
A6.13.6.1.4a, A6.13.6.1.4b and A6.13.6.1.4c are revised
B.6 - 60
September 2007
SPECIFICATIONS
COMMENTARY
below to give two sets of provisions. The straight girder
provisions correspond to the AASHTO-LRFD Bridge
Design Specifications, Second Edition, 1998. The
horizontally curved girder provisions correspond to the
AASHTO-LRFD Bridge Design Specifications, Third
Edition, 2004. Both were revised for PennDOT use.
Using the 1998 AASHTO-LRFD as basis for the design of
straight girder splices allows the Department to continue
using the existing steel splice design computer program
(SPLRFD). It is the intent of the Department to update
the SPLRFD computer program in the future to
correspond to the provisions of the third edition of the
AASHTO-LRFD Specifications.
6.13.6.1.4a General
A6.13.6.1.4a shall be revised as follows:
For straight bridges:
The following shall replace A6.13.6.1.4a.
Splice plates shall be investigated for fatigue of the base
metal adjacent to slip-critical connections and specified in
Table A6.6.1.2.3-1, using the gross section of the splice
plates and member.
Splices subjected to tension shall satisfy the requirements
specified in A6.13.5.2.
For horizontally curved bridges:
The following shall replace the fourth paragraph of
A6.13.6.1.4a.
The factored flexural resistance of the flanges at the point
of splice at the strength limit state shall satisfy the
applicable provisions of A6.10.6.2, D6.10.6.2.
6.13.6.1.4b Web Splices
C6.13.6.1.4b
A6.13.6.1.4b shall be revised as follows:
A6.13.6.1.4b shall be revised as follows:
The following shall replace A6.13.6.1.4b.
The elastic methods shall be used to determine the shear
force in splice bolts. The ultimate strength method is not
permitted.
Web splice plates and their connections shall be designed
at the strength limit state for:
The following shall replace AC6.13.6.1.4b.
For bolt groups subjected to eccentric shear, a traditional
conservative approach is often used in which the bolt
group is treated as an elastic cross-section subjected to
direct shear and torsion. A vector analysis is performed
assuming that there is no friction and that the plates are
rigid and the bolts are elastic.
The portion of the factored design moment, specified
in A6.13.1 and D6.13.1, that is resisted by the web;
The moment due to eccentricity of a notional shear
determined as the shear due to the factored loading
multiplied by the design moment specified in A6.13.1
and D6.13.1 and divided by the moment caused by
the factored loads and the shear itself; and
B.6 - 61
SPECIFICATIONS
September 2007
COMMENTARY
The notional shear itself.
At the strength limit state, the flexural stress in the splice
plates shall not exceed the specified minimum yield
strength of the splice plates.
Web splice bolts shall be designed for the effects of
moment due to the eccentric shear.
Web plates shall be spliced symmetrically by plates on
each side. The splice plates for shear shall extend the full
depth of the girder between flanges. There shall be not
less than two rows of bolt on each side of the joint.
For bolted web splices with thickness differences of 2 mm
{0.0625 in.} or less, no filler plates are required.
The following shall supplement A6.13.6.1.4b.
The elastic methods shall be used to determine the shear
force in splice bolts. The ultimate strength method is not
permitted.
The following shall replace the second sentence of the
first paragraph of A6.13.6.1.4b.
For all single box sections, and for multiple box sections
in bridges not satisfying the requirements of A6.11.2.3,
including horizontally curved bridges, or with box flanges
that are not fully effective according to the provisions of
A6.11.1.1 and D6.11.1.1, the shear shall be taken as the
sum of the flexural and St. Venant torsional shears in the
web subjected to additive shears.
6.13.6.1.4c Flange Splices
C6.13.6.1.4c
A6.13.6.1.4c shall be revised as follows:
AC6.13.6.1.4c shall be revised as follows:
The following shall replace A6.13.6.1.4c.
At the strength limit state, the axial stress in the flange
splice plate shall satisfy the requirements of A6.13.5.2 if
in tension and A6.9.2 and D6.9.2 if compression.
For bolted flexural members, bolted splices in flange parts
should not be used between field splices, unless approved
by the Engineer. In any one flange, not more than one
part should be spliced at the same cross-section. If
practicable, splices should be located at points where
there is an excess of section.
The following shall replace A6.13.6.1.4c.
For compression, use an unbraced length equal to zero,
making Pn = Fy As
paragraph before last of A6.13.6.1.4c.
For all single box sections, and for multiple box sections
in bridges not satisfying the requirements of A6.11.2.3,
including horizontally curved bridges, or with box flanges
that are not fully effective according to the provisions of
A6.11.1.1 and D6.11.1.1, longitudinal warping stresses
due to cross-section distortion shall be considered when
checking bolted flange splices for slip and for fatigue.
The following shall replace the first sentence of the tenth
paragraph of AC6.10.6.1.4c.
For the box sections cited in this article, including
sections in horizontally curved bridges, longitudinal
warping stresses due to cross-section distortion can be
significant under construction and service conditions and
must therefore be considered when checking the
connections of bolted flange splices for slip and for
fatigue.
B.6 - 62
SPECIFICATIONS
September 2007
COMMENTARY
eleventh paragraph of AC6.10.6.1.4c.
In cases for straight girders where flange lateral bending is
deemed significant, and for horizontally curved girders,
the effects of the lateral bending must be considered in the
design of the bolted splices for discretely braced top
flanges of tub sections or discretely braced flanges of Isections.
6.13.6.1.5 Fillers
A6.13.6.1.5.
Fillers 6.0 mm {1/4 in.} or more in thickness shall consist
of not more than two plates, unless approved by the Chief
Bridge Engineer.
The following shall replace the fifth paragraph of
A6.13.6.1.5.
The specified minimum yield strength of fillers 6.0 mm
{1/4 in.}or greater in thickness shall not be less than the
larger of 70 percent of the specified minimum yield
strength of the connected plate and 250 MPa {36 psi}
unless approved by the Chief Bridge Engineer.
6.13.6.2 WELDED SPLICES
C6.13.6.2
A6.13.6.2.
Welded field splices shall not be used without written
Use the AASHTO/AWS D1.1M/D1.1:2002 Structural
Welding Code for the welding of new tubular structures,
pipes, piles, and existing which are not covered by
AASHTO/AWS D1.5M/D1.5:2002.
6.15 PILES
6.15.1 General
C6.15.1
The following shall replace A 6.15.1.
Piles shall be designed as structural members capable of
safely supporting all imposed loads.
For a pile group composed of only vertical piles which is
subjected to lateral load, the pile structural analysis shall
include explicit consideration of soil-structure interaction
effects using a COM624P analysis (Wang and Reese,
1993) or LPILE 5.0 analysis.
Based on the parametric study conducted by the
Department, which is described in the commentary, an
abutment or retaining wall with a pile group composed of
both vertical and battered piles which is subjected to
lateral load shall be designed assuming that all lateral load
is resisted by the horizontal component of the axial
The following shall replace AC6.15.1.
To develop the recommended distribution of lateral load
among piles supporting a typical bridge abutment, a
parametric study (Kelly, et al, 1995) was performed using
the program GROUP (Reese, et al, 1994). A second
purpose of this parametric study was to determine if the
Department's lateral deformation criteria of 12 mm {1/2
in.} for the service limit state and 25 mm {1 in.} for the
strength limit state were satisfied. These criteria were met
for all analyses representative of the Department's
practice. The variables evaluated in the parametric study
included:
HP310X79 {HP12X53} and HP250X62 {HP10X42}
B.6 - 63
SPECIFICATIONS
September 2007
COMMENTARY
capacity of the battered piles. The vertical load shall be
distributed among piles in the group using a simple elastic
procedure. The use of the above design procedure with
any of the following conditions requires the approval of
the Chief Bridge Engineer:
piles with about four-pile diameters center-to-center
spacing;
End bearing piles on rock driven through a medium
stiff-to-stiff clay deposit, and friction piles in deposits
of medium dense and loose sand;
Piles with a specified steel yield strength other than
250 MPa {36 ksi}
Pile lengths of 3.0, 9.1 and 15.2 m {10, 30, and 50
ft.};
Piles with bending stiffness properties less than
HP250X62 {HP10X42}
Vertical and horizontal load levels consistent with
common Department designs; and
Very soft clays or very loose sands as defined in
Standard Drawing BC-795M
Pile-head fixity conditions of fixed, pinned and
partially fixed.
Piles with bending stiffness properties less than
HP310X79 {HP12X53} in soft clays or loose sands
as defined in Standard Drawing BC-795M
For typical pile groups containing battered piles
designed using the simplified procedure, the pile
study indicates that:
Unfactored vertical load to horizontal load ratio less
than 3.5 (excluding seismic forces)
Combined stresses in upper portions of the battered
and vertical piles due to axial load and bending are
generally less than 82 to 95 MPa {11.9 to 13.8 ksi},
which is consistent with previous practice.
The fraction of the total lateral load resisted by
bending of the vertical piles is generally less than
about 20 percent.
The check of structural pile capacity for combined
axial load and flexure in the upper portion of the pile
using the LRFD Interaction Equations in A6.9.2.2
does not control the pile design.
Lateral deflections are well below acceptable
magnitudes.
As pile stiffness increases, horizontal deformations
and associated bending stresses decrease such that the
simplified method remains applicable.
In cases for which pile and soil conditions differ
significantly from those conditions examined in the
parametric study, a suitable analysis should be performed
which incorporates the necessary soil-structure interaction
factors. This analysis may comprise finite element
analysis, p-y analysis, or other applicable methods.
Lateral deflections and maximum bending stresses for
laterally-loaded pile groups generally occur within a depth
below the pile cap equal to approximately 10 pile
diameters. Therefore, the presence of poor material (very
soft clays or very loose sands) within the upper 10 pile
diameters invalidates use of the simplified method due to
the potential for pile overstressing and excessive
deformations under lateral loads. If these conditions exist,
the designer may consider the following options:
B.6 - 64
SPECIFICATIONS
September 2007
COMMENTARY
Improve in place or remove and replace the poor
material. These may be viable options when the
thickness of poor material is small and close to the
ground surface.
Perform a more rigorous, problem specific analysis to
define pile stress levels and pile group deformations.
This type of analysis may be performed using
software such as GROUP, Reese, et al (1994).
A thin seam or lens of poor material below the upper 10
pile diameters will not typically affect the applicability of
the simplified method.
6.15.2 Structural Resistance
C6.15.2
Resistance factors, φ, for the strength limit state shall be
taken as specified in D6.5.4.2. The resistance factors for
axial resistance of piles in compression which are subject
to damage due to driving shall be applied only to that
section of the pile likely to experience damage.
Therefore, the specified φ factors of 0.35 and 0.45,
specified in D6.5.4.2 for piles subject to damage, shall be
applied only to the axial capacity of the pile. In addition,
the φ factors of 0.60 and 0.85, specified in D6.5.4.2 for
axial and flexural resistance, respectively, of undamaged
piles, shall be applied to the combined axial and flexural
resistance of the pile in the interaction equation for the
compression and flexure terms respectively. This design
approach is illustrated on Figure 1.
For piles bearing on soluble bedrock, the φ factor of 0.25
shall be applied to the axial capacity of the pile using a
pile yield strength Fy = 250 MPa {36 ksi}.
Due to the nature of pile driving, additional factors must
be considered in selection of resistance factors that are not
normally accounted for in steel members. The factors
considered in development of the specified resistance
factors include:
Unintended eccentricity of applied load about
expected point of application,
Variations in material properties of pile, and
Pile damage due to driving.
These factors are discussed by Davisson (1983). While
the resistance factors specified herein generally conform
to the recommendations given by Davisson (1983), they
have been modified to reflect the common practice of the
Department. Specifically, the resistance factors presented
in D6.5.4.2 have been selected in a manner such that,
when combined with an average load factor of 1.45, the
equivalent factor of safety calculated as the ratio of the
appropriate load to resistance factor is comparable to the
factor of safety previously used by the Department.
The factored compressive resistance, Pr, includes
reduction factors for unintended load eccentricity and
material property variations, as well as a reduction for
potential damage to piles due to driving, which is most
likely to occur near the tip of the pile. The resistance
factors for computation of the factored axial pile capacity
near the tip of the pile are 0.35 and 0.45 for severe and
good driving conditions, respectively. The φ factor of
0.25 for piles bearing on soluble bedrock is intended to
safeguard against the potential of the loss of geotechnical
capacity in soluble bedrock.
For steel piles, flexure occurs primarily toward the head
of the pile. This upper zone of the pile is less likely to
experience damage due to driving. Therefore, relative to
combined axial compression and flexure, the resistance
B.6 - 65
September 2007
SPECIFICATIONS
COMMENTARY
factor for axial resistance (φc = 0.60) accounts for both
unintended load eccentricity and pile material property
variations, whereas the resistance factor for flexural
resistance (φf = 0.85) accounts only for variations in pile
material properties.
Typically, due to the lack of a detailed soil-structure
interaction analysis of pile groups containing both vertical
and battered piles, evaluation of combined axial and
flexural loading will only be applied to pile groups
containing no battered piles.
Figure 6.15.2-1 - Distribution of Moment and Deflection
in Vertical Piles Subjected to Lateral Load
6.15.3P Compressive Resistance
The design of steel piles shall follow A6.9, except as
specified herein.
6.15.3.1 AXIAL COMPRESSION
The following shall replace A 6.15.3.1.
For piles under axial load, the factored resistance of piles
in compression, Pr, shall be taken as specified in A6.9.2.1
using the resistance factor, φc, specified in D6.5.4.2.
B.6 - 66
SPECIFICATIONS
6.15.3.2 COMBINED AXIAL COMPRESSION AND
FLEXURE
The following shall replace A 6.15.3.2.
Piles subjected to axial load and flexure shall be designed
in accordance with A6.9.2.2 using the resistance factors,
φc and φf, specified in D6.5.4.2.
Vertical H-pile foundations designed using COM624P or
LPILE 5.0 per D10.7.3.12.2 may use the values given in
Tables 1 and 2.
where:
D
=
Depth of the pile.
Area
=
Area of the pile.
Ix, Iy
=
Moment of inertia about their respective
axis.
PuSERV
=
COM624P or LPILE 5.0 pile load
equivalent to pile resistance at Service Limit
State.
PuSTR
=
COM624P or LPILE 5.0 pile load
equivalent to pile resistance under severe
driving conditions as defined in D6.15.2P.
Pr
=
Factored axial resistance for combined axial
and flexural resistance.
Mrx, Mry = Factored flexural resistance of the vertical
pile in the x-axis and y-axis, respectively.
September 2007
COMMENTARY
C6.15.3.2
The factored flexural resistance, Mrx, is based on either the
plastic or elastic moment of pile considering web and
compression flange slenderness requirements. The
factored flexural resistance, M ry, is based on the plastic
moment per AC6.12.2.2.1.
For these tables the piles are considered as braced. If very
weak soils, scour or voids are expected, the buckling
requirements of DM-4 6.15.3.3 must be considered and
the values shown in Tables 1and 2 are not applicable.
PuSERV and PuSTR in Tables 1 and 2 are based on 0.25*fy*As
and 0.35*fy*As, respectively. For piles bearing on
soluble rock (limestone,etc.) values for P uSERV and PuSTR
equal to 0.16*fy*As and 0.25*fy*As, respectively, are to
be considered.
The section properties provided are for use in the
COM624P or LPILE 5.0 analysis. The combined axial
compression and flexural requirements of A6.9.2.2 shall
be evaluated considering the results of the COM624P or
LPILE 5.0 analysis and the resistances provided in the
Tables 1 and 2.
The section properties may also be used in PENNDOT’s
Integral Abutment Spreadsheet. However; since the
capacities in the tables do not consider unbraced length,
the structural pile capacity of vertical piles used in an
Integral Abutment must be checked in accordance with
DM-4 Appendix G using the Integral Abutment
Spreadsheet.
B.6 - 67
September 2007
SPECIFICATIONS
COMMENTARY
Table 6.15.3.2P-1 – Pile Properties, Factored Axial and Flexural Resistances with Full Pile Section.
Fy = 250 MPa {36 ksi}
Metric
HP
Depth
D (mm)
Area
(mm2)
Ix
(106mm4)
Iy
(106mm4)
PuSERV
(kN)
PuSTR
(kN)
Pr
(kN)
Mrx
(kN-m)
Mry
(kN-m)
360 x 174
361
22200
508
184
1378
1929
3306
679
310
360 x 152
356
19400
439
159
1204
1685
2889
589
270
360 x 132
351
16900
375
135
1049
1468
2517
455
231
360 x 108
346
13800
303
108
856
1199
2055
372
186
310 x 125
312
15900
270
88.2
987
1381
2368
417
180
310 x 110
308
14100
237
77.1
875
1225
2100
368
158
310 x 94
303
11900
196
63.9
738
1034
1772
274
132
310 x 79
299
10000
163
52.6
621
869
1489
232
110
250 x 85
254
10800
123
42.3
670
938
1608
232
104
250 x 62
246
7970
87.5
30.0
495
692
1187
151
75
US Customary Units
HP
Depth
D (in)
Area
(in2)
Ix
(in4)
Iy
(in4)
PuSERV
(kips)
PuSTR
(kips)
Pr
(kips)
Mrx
(kip-ft)
Mry
(kip-ft)
14 x 117
14.21
34.4
1220
443
310
434
743
495
228
14 x 102
14.01
30.0
1050
380
270
378
648
431
197
14 x 89
13.83
26.1
904
326
235
329
564
334
169
14 x 73
13.61
21.4
729
261
193
270
462
273
137
12 x 84
12.28
24.6
650
213
221
310
531
306
132
12 x 74
12.13
21.8
569
186
196
275
471
268
116
12 x 63
11.94
18.4
472
153
166
232
397
202
97
12 x 53
11.78
15.5
393
127
140
195
335
170
81
10 x 57
9.99
16.8
294
101
151
212
363
170
75
10 x 42
9.70
12.4
210
71.7
112
156
268
111
54
B.6 - 68
September 2007
SPECIFICATIONS
COMMENTARY
Table 6.15.3.2P-2 – Factored Axial and Flexural Resistance with 1.5 mm (1/16") Section Loss. F y = 250 MPa (36 ksi)
Metric
HP
Depth D
(mm)
Area
(mm2)
Ix
(106mm4
)
Iy
(106mm4
)
PuSERV
(kN)
PuSTR
(kN)
Pr
(kN)
Mrx
(kN-m)
Mry
(kN-m)
360 x 174
358
18530
424
152
1150
1610
2760
504
258
360 x 152
353
15730
355
127
976
1367
2343
428
217
360 x 132
348
13240
295
105
821
1150
1971
360
181
360 x 108
343
10210
224
79.7
634
887
1520
278
138
310 x 125
309
12780
217
70.1
793
1110
1903
299
145
310 x 110
305
10980
184
59.5
682
954
1636
257
124
310 x 94
300
8820
145
46.8
547
766
1313
206
98
310 x 79
296
6980
113
36.4
433
606
1039
163
77
250 x 85
251
8290
93.3
31.5
514
720
1,234
158
78
250 x 62
243
5470
59.9
20.2
339
475
814
105
51
US Customary Units
HP
Depth D
(in)
Area
(in2)
Ix
(in4)
Iy
(in4)
PuSERV
(kips)
PuSTR
(kips)
Pr
(kips)
Mrx
(kip-ft)
Mry
(kip-ft)
14 x 117
14.09
28.73
1019
365
259
362
620
369
189
14 x 102
13.89
24.39
853
305
219
307
527
313
159
14 x 89
13.71
20.51
708
253
185
258
443
263
133
14 x 73
13.49
15.83
537
192
142
199
342
203
101
12 x 84
12.16
19.81
521
168
178
250
428
219
106
12 x 74
12.01
17.02
443
143
153
215
368
188
90
12 x 63
11.82
13.66
349
112
123
172
295
151
72
12 x 53
11.66
10.81
272
87.5
97
136
234
119
56
10 x 57
9.87
12.84
224
75.6
116
162
277
116
57
10 x 42
9.58
8.48
143
48.5
76
107
183
77
37
B.6 - 69
September 2007
SPECIFICATIONS
COMMENTARY
6.15.3.3 BUCKLING
C6.15.3.3
The following shall replace the last sentence of A 6.15.3.3.
The depth to fixity shall be determined in accordance with
D10.7.3.13.4 for battered piles or COM624P or LPILE 5.0
analyses for vertical piles. Figure D6.15.2-1 illustrates the
depth to fixity as determined by COM624P or LPILE 5.0.
The following shall replace AC 6.15.3.3.
The use of an approximate method in lieu of a Panalysis is allowed if approved by the Chief Bridge
Engineer.
6.15.4 Maximum Permissible Driving Stresses
Maximum permissible driving stresses for top driven steel
piles shall be taken as specified in D10.7.8.
B.6 - 70
September 2007
SPECIFICATIONS
COMMENTARY
APPENDIX A – FLEXURAL RESISTANCE – COMPOSITE SECTIONS IN NEGATIVE FLEXURE AND
NONCOMPOSITE SECTIONS WITH COMPACT OR NONCOMPACT WEBS
The following shall be added immediately after the heading of
the appendix and before AA6.1.
The provisions of this Appendix are applicable to curved
bridge components designed using the provisions of A6.10
and A6.11 as revised by D6.10 and D6.11. They are not
applicable to straight bridge components designed using the
provisions of Appendix DE.
APPENDIX B – MOMENT REDISTRIBUTION FROM INTERIOR-PIER SECTIONS IN CONTINUOUS-SPAN
BRIDGES
Delete Appendix B in its entirety
The provisions of Appendix B correspond to the inelastic
design procedures that are not allowed in Pennsylvania.
APPENDIX C – BASIC STEPS FOR STEEL BRIDGE SUPERSTRUCTURES
The following shall be added immediately after the heading of
the appendix and before AC6.1.
The provisions of this Appendix are applicable to curved
bridge components designed using the provisions of A6.10
and A6.11 as revised by D6.10 and D6.11. They are not
applicable to straight bridge components designed using the
provisions of Appendix DE.
B.6 - 71
September 2007
REFERENCES
The following shall supplement the references of A5.
Albrecht, P., Coburn, S.K., Wattar, F.M., Tinklenberg, G.L. and W.P. Gallagher. Guidelines for the Use of Weathering Steel
in Bridges. NCHRP Report 314. TRB, National Research Council, Washington, D.C., June 1989.
Mertz, D. R., "Displacement-Induced Fatigue Cracking in Welded Steel Bridges", Ph.D. dissertation, Lehigh University,
1984
B.6 - 72

Guidelines for Analyzing Curved and Skewed Bridges

Transcription

Similar documents

info! - American Senior Communities

Challenging Suspension Bridges

Salginatobel Bridge

Suspension Wire 2

Designing for Increased Traffic Loading

Community Bridges Flyer:Layout 1

Kids Trail - Train World

Centurion Mk 5 Bridgelayer

BRAZOS RIVER BRIDGES